The effect of mergers and acquisitions on the dividend policy of banks Nnadi, M.A. Submitted version deposited in CURVE May 2011 Original citation: Nnadi, M.A. (2010). The effect of mergers and acquisitions on the dividend policy of banks. Unpublished PhD Thesis. Coventry: Coventry University. Copyright © and Moral Rights are retained by the author. A copy can be downloaded for personal non-commercial research or study, without prior permission or charge. This item cannot be reproduced or quoted extensively from without first obtaining permission in writing from the copyright holder(s). The content must not be changed in any way or sold commercially in any format or medium without the formal permission of the copyright holders. One table only on page 84 has been removed for copyright reasons. The unabridged version of the thesis can be viewed at the Lanchester Library, Coventry University.

CURVE is the Institutional Repository for Coventry University http://curve.coventry.ac.uk/open

iii

The Effect of Mergers & Acquisitions

on the Dividend Policy of Banks

M.A. NNADI

A thesis submitted in partial fulfilment of the Coventry University’s

requirements for the Degree of Doctor of Philosophy (PhD)

2010

iv

Table of Contents

Supervisory team---------------------------------------------------------------------ii

Table of Contents--------------------------------------------------------------------iii

List of Tables & Figures------------------------------------------------------------iv

List of Abbreviations ---------------------------------------------------------------v

List of Appendices -------------------------------------------------------------------vi

Abstract --------------------------------------------------------------------------------vii

Acknowledgement-------------------------------------------------------------------viii

Dedication ------------------------------------------------------------------------------ix

Chapter 1 Introduction ---------------------------------------------------------------------1

Chapter 2 Review of Literature on Dividend Policy ------------------------------------12

Chapter 3 Theoretical Debates & Dividend Policy Hypotheses ----------------------31

Chapter 4 Analyses of Mergers & Acquisitions -----------------------------------------74

Chapter 5 Methodology ---------------------------------------------------------------------118

Chapter 6 Data Analyses & Discussion of Result --------------------------------------137

Chapter 7 Factors Influencing Bank Dividend Policy & M&A Effect ------------157

Chapter 8 Summary of Major Findings & Concluding Remarks ------------------177

References --------------------------------------------------------------------------------------185

Appendices

v

List of tables & figures

Table 1.1 Some M&A Studies and the Length of the windows

Table 4.1 Trend in the Number of Banks Among EU Countries

Table 6.1 Research Hypotheses, Proxies and their Expected Outcomes.

Table 6.2 The Cumulative Total Abnormal Returns of Cross border & Domestic M&A

Table 6.3 Cumulative Totals Abnormal Returns in Quartiles

Table 6.4 Summary Statistics of the Abnormal Returns and Merger Performance of the

Domestic and Cross Border Banks

Table 6.5 Correlations of CTSAR with Pre- Domestic Performance

Table 6.6 Correlations of CTSAR with Post Domestic Performance

Table 6.7 Model Summary of the Hierarchical Regression Financial Performance of the

Domestic Bank Acquirers

Table 6.8 Correlations of CTSAR with Pre Cross border Performance

Table 6.9 Correlations of CTSAR with Post Cross border Performance

Table 6.11 Model Summary of Financial Performance of the Cross Border Bank Acquirers

Table 7.1 Descriptive Statistics on Measures of Dividend Policy for Domestic &

Cross border M&A (1997-2007)

Table 7.2 Univariate Statistics of Measures and Factors Affecting Dividend Policy

Table 7.3 Model Summary of the Hierarchical Regression

Table 7.4 Granger Causality Test Result

Figure 3.1 Graphical representation of the study time frame

Figure 5.1 Graph of the Cross border Cumulative Total Standardized Abnormal Returns

Figure 5.2 Graph of the Domestic Cumulative Total Standardized Abnormal Returns

vi

List of Abbreviations

CapFnSt = capital and finance structures measured by the Debt/Equity ratio (capital

structure) while the finance structure is debt + equity/total assets;

CIR = Cost to income ratio

CTSAR = Cumulative total standardised abnormal returns

EPS = Earnings per share

EQTASS = Equity to total assets

LOANoDEP = Loans over deposits to show the percentage of deposits that are

returned to the banks through loans

LOANTAS = Net loans to total assets

NII = Net interest income

PROVTAS = Loan loss provision to total assets

ROE = Return on equity

SIZE = natural logarithm of the total assets

EU = European Union

EC = European Commission

vii

List of Appendices

1. List of M&A cross border acquirers

2. List of M&A domestic acquirers

viii

Abstract

The number of domestic and cross border bank mergers and acquisitions (M&A) has

increased over the last decade with a resultant impact on the bank dividend. This study

examines the effect of M&A on the dividend policy by comparing the abnormal returns,

profitability and dividend policy of the domestic and cross border bank acquirers. The study

focused on EU mega-bank mergers and acquisitions within 1997-2007 involving only

commercial-to-commercial banks. The sample consists of a total of 62 mega-M&A with a

minimum deal value of €500 million. Three hypotheses were formulated specifically to test:

(i) the wealth effect and geographical diversification of the M&A between domestic and

cross border acquirers; (ii) the effect on in the financial performance of both acquirers and

(iii) the M&A impact on dividend policy on banks after bank M&A. Two strands of the

literature were reviewed focusing on M&A and dividend policy.

The event study methodology was used to calculate the abnormal returns of both the domestic

and cross border acquirers which were standardised. A long window of 61 days was applied

to capture a satisfactory length of pre and post merger events that could capture the behaviour

of the abnormal returns and consequent effect on dividend policy. The hierarchical

regression model was used to estimate the impact of the variables on the profitability and

dividend policy of the acquirer banks.

In comparison with the domestic acquirers, the cross border abnormal returns showed a trend

of significant negative results following the M&A announcement. The domestic acquirers

showed no significance but, on average have higher cumulative total standardised abnormal

returns (CTSAR) than the cross border acquired banks. The result of the financial

performance showed that CTSAR of the cross border acquirers is significantly affected by the

profitability of the banks but insignificant with domestic acquirers. However, the cost-to –

income ratio (CIR) significantly affects the performance of both bank acquirers. CIR and

RISK (measured by the ratio of the loan provision to net interest revenue of the banks) highly

correlated with profitability of both the domestic and cross border acquirers. The

management of costs and loans risks were found to be significant variables in the

achievement of profitability among domestic acquirers.

The dividend policy hypothesis result indicated that CTSAR has a weak correlation and

insignificant effect on the dividend policy variables. Infact, the Causality test result

confirmed that the CTSAR does not Granger cause dividend policy. However, the study

provides strong support to previous studies that beta, liquidity, taxes, and the finance

structure of the acquirers are significant variables in the formulation of the dividend policy of

the merged banks. The beta, which a proxy for risk, is the most significant factor affecting

the dividend policy of the merged banks.

Acknowledgement

The dream of this programme was conceived in the wilderness, hatched in the desert but

completed in the spring of summer by Divine providence! I thank Jehovah God for His

abundant grace which has been generously extended to me, on the basis of which this

programme was completed.

I am grateful to my Director of Studies, Dr. Sailesh Tanna for his supervision, interest and

invaluable assistance during the programme. I also thank Dr Fotios Pousiouras for his kind

contribution especially with the download of the most of the datasets used in the research.

I wish to also thank my examiners; Professor Phil Molyneux and Mr. Keith Redhead for their

corrections and suggestions on improving the quality of the thesis.

My commitment to finishing this programme has in no doubt tested the patience of my

family. I am immensely grateful to Josephine and Chidinma for their support and patience at

such times when I kept away.

I am grateful to my parents, Mr. & Mrs Ishmael Nnadi Nweke for their inspiration and

prayers that helped to sustain me through these years. My siblings Peace, Chidiebere, Joseph,

Okwuchi, Nnenna and Emeka have been of immerse support. I also thank Mr. & Mrs Gospel

Nweke (my parents in-law); Princewill, Joanah and Osinachi for their company and support

during these ‘long years’ of the study.

I wish to thank all my senior colleagues at the Rivers State University of Science &

Technology, Port Harcourt, Nigeria for their fervent moral support and encouragement;

special thanks to Dr ( Mrs) Meg Akpomi for her extra ordinary concern and visits during the

programme. I also thank my dean, Professor Wey Amaewhule for his moral encouragements

as well as Dr (Mrs) Maureen Koko. I passionately thank all my friends who have helped me

this date; Promise Ogundu Esq., Royal Ayebatari Wilson & family, Christian Onigwe, Isaac

Ogundu, Chika Woka & family and my Christian brothers & sisters for their spiritual support

and encouragement. Thanks to Missi and Ms Umosen for their assistance in proofreading my

work.

Finally, I owe a debt of gratitude to the Chartered Institute of Taxation (CITA) for their

financial grant towards the completion of my thesis.

x

Dedication

To my grandma

Madam Christiana Azuonwu

& my daughter

Chidinma Nnadi

1

Chapter One

Introduction

1.0 Background of the Study

The issue of corporate dividend policy has dominated finance studies over the decades. The

interest in dividend theories of organisations by scholars can be justified by the crucial role of

dividend in investment and financing decisions. Such decisions are hardly made on the whim

particularly by banks. The banks, like other organisations, are usually faced with the

challenge of what decisions to make with generated profits of the organisation. The decision

has to be made between investing all or proportion of the earned profits in the form of

dividends to the shareholders, or to plough it back into the business. The best decision

however is one that maximises the wealth of the shareholders and exerts a positive impact on

its share price.

The dividend policy decision of a bank is anchored on a number of factors such as; liquidity,

the composition of the bank, the tax implications, the growth and future investments of the

bank, the capital structure etc. Most corporate organisations realise the role of dividend

payments in satisfying shareholders expectations. In fact, dividend decision has been

intrinsically linked with the improvement of shareholders wealth. The 1960‟s witnessed the

surge on the dividend policy debates among finance scholars which was pioneered by

Modigliani and Miller (1961). The theory, also known as M&M, among other issue posits

that dividend was irrelevant in increasing the value of a firm. Several other theories had

since then flooded the finance literature either in support or in opposition of the theory. Some

others have even taken a different view, thus making dividend policy an ongoing issue.

The foregoing argument underscores the overwhelming sensitivity of dividend decisions on

the image and corporate existence of organisations. As a result, any strategic alliance or

measure to consolidate dividends with the aim of improving the shareholders wealth by banks

would be ultimately considered. Most banks have considered mergers and acquisition (M&A)

as a strategic option, not only for expansion but also to improve the returns accruable to their

shareholders. In the European Union countries, M&A deals have become a common trend

2

particularly during the period 1997 -2007. While different banks may have various missions

in pursuing a merger strategy, the ultimate aim is to improve the returns to shareholders and

therefore strengthen their dividend policy. This study attempts to link the issue of M&A

announcements to the dividend policy and profitability of domestic and cross border

commercial bank acquirers.

A huge number of studies in the realm of bank mergers and acquisitions (M&A) have

focused on the returns accruable to the target versus acquirer shareholders. An overwhelming

majority of the studies confirm that the shareholders of the targets benefit more than those of

acquirers from M&A activities. The evidence is often based on event studies with a limited or

short window around the announcement date and using the banks and market returns to

estimate the abnormal returns to the target or the acquirer shareholders.

While these studies have produced different results, most of them have given less attention to

the geographical implications of the M&A deals on the shareholders, particularly of the

acquirers. Since the general notion is that the acquirers‟ shareholders have little or no benefits

in a M&A, dissecting the returns into cross border and domestic shareholders would be

insightful in determining the underlying factors why the acquirer shareholders benefit less.

The focus of this study is therefore on the dividend policy implications of the M&A on the

acquirer banks, distinguishing between cross border and domestic bank acquires.

It is important to state that M&A studies on cross border and domestic M&A are quite vast.

However, most of the studies are limited by short event windows which do not capture

significant events and performance of the merged entities before and after the event. They

have often focused on the abnormal returns to the target banks eg Rad & Beek, 1999; Cybo-

Ottone & Murgia, 2000; Kane, 2000. This study expands on the event window and uses a

long estimation period to capture the „behind the scene‟ performance of the domestic and

cross border entities and seeking to obtain a clue as to why the acquirer shareholders gain less

in M&A. The 61 days used in the study is considered sufficient and appropriate to make

deductive observations which short event windows are unlikely to capture. Becker (2000)

confirms that the choice of window is very crucial to the expectant result of the returns

particularly for the acquirers which could range from statistically negative to insignificantly

positive.

3

In addition, geographical factors could reveal vital causes on why the acquirer shareholders

returns are often in the negative. Environmental factors such as country‟s legislations, attitude

of citizens towards foreign acquisitions, knowledge of the local market, size and capital can

have a considerable effect on the acquirer banks in cross border deals.

The methodology in most studies has followed the traditional Capital Assets Pricing Model

(CAPM) and the application of the market model in the estimation of the cumulative

abnormal returns (CARs) to measure shareholders returns. This study makes a departure from

existing studies by using the cumulative total standardized abnormal returns (CTSAR) to

cater for the different degree of event impact rather than using the traditional CARs. This is

done by weighing each abnormal return by the standard deviation. The purpose of the

standardization is to ensure that each abnormal return has the same variance (Serra, 2002).

Thus, by dividing each firm‟s abnormal residual by the standard deviation over the estimation

period, each residual has an estimated variance of 1. Rarely have studies in cross border and

domestic M&A applied this technique in combination of a long window to determine M&A

effect on bank dividends. The method also helps to mitigate any bias in the abnormal returns

of both acquirers. Cornett & Tehranian (1992) applied the standardized abnormal returns

technique but in a 2 day window, to measure the post-merger performance of 30 large bank

acquisitions with the pre-merger performance of the merging banks.

Shareholders‟ wealth and returns should be seen as a generic concept rather than being

limited to the increase and decrease in the abnormal returns. The use of a single variable in

the measurement and estimation of shareholders wealth and abnormal returns may not be

sufficient in analysing the robustness of results. Thus, in addition to the abnormal returns,

this study also uses the dividend payout as a measure of M&A gains to the acquirers‟

shareholders.

Although studies in dividend policy have invoked contentious debates in finance, there is a

dearth of research into its relationship with M&A. In fact, no known study has examined the

effect of M&A on dividend policy particularly in the EU banking industry. Studies such as

Olson & Pagano (2005), Page, Jahera & Pugh (1996), Rozeff (1982), Brook, Hendershott &

Lee (1998), Casey & Dicken (2000), Johnson et al (2006) and Pattenden & Twite (2008)

have considered various facets of dividend policy both for banking and other industries. This

study is unique in that it examines the issue of dividend policy among cross border and

4

domestic acquirers. It assesses the impact of profitability on dividend policy and uses the

CTSAR in combination with other variables to evaluate the factors affecting banks dividend

policy. The concept of dividend policy in the present study is measured in terms of the

dividend yield and payout. The term can also be applied to include either one or more of the

two elements (yield and payout) and others, such as the earnings per share (EPS)1.

Dividend yield and payout have been interchangeably used in various dividend policy studies

though the yield offers more significant result. This is not surprising as the concept is even

more encompassing. The dividend payout ratio represents the proportion of the earnings that

the banks pay out to shareholders in the form of dividends. It is usually the net profit after

taxation and after the preference dividends announced during the period; expressed as a

percentage. The dividend yield relates the cash return from a share to its current market value

which is often stated at the grossed up rate (Reuters, 2008).

1.1 Previous Studies

Studies in dividend policy including Rozeff (1982); Casey & Dickens (2000); La Porta (2000);

Fama & French (2001); Grullon et al (2003); Gugler & Yurtoglu (2003); Brunarski, Harman &

Kehr (2004); Johnson et al (2006) have analyzed the various aspects of dividend policy but

none have examined banks dividend policy in relation to the abnormal returns to the

shareholders of cross border and domestic bank acquirers in mergers and acquisitions.

Rozeff (1982) and Casey & Dickens (2000) examined the effects of tax and regulatory changes

on commercial banks dividend policy. Both studies used only dividend payout as a measure of

dividend policy and applied the value line beta as proxy for bank leverage. They found that the

estimation of the leverage by value line can be precarious as the bank leverage ratio would

merely provide better picture of its level of debts.

Gugler & Yurtoglu (2003) applied the cumulative average abnormal returns (CAAR) to

evaluate the structure of ownership of majority controlled firms in Germany using the rent

1 See chapter three for the elaborate concept of dividend policy. However, the above three measures (yield,

payout and EPS) are commonly used in part or jointly to describe dividend policy. Chen et al (2005) used

payout and yield; Gugler and Yurtoglu (2003) used payout; while La- Porta et al (2000) used a series of

dividend ratios such as dividend to sales, dividend to earning etc as instruments of measuring a firm‟s dividend

policy.

5

extraction hypothesis. But the study does not focus on cross border cum domestic acquires; in

particular a window of -/+ 2days was used which may be too short to capture the long run

performance of the firms. Furthermore, the study only concentrated on the dividend payout and

does not consider the yield.

In an international setting, La Porta, et al (2000) examined the substitute and outcome dividend

models for a cross section of 4000 companies from 33 countries and upheld that the outcome

model is more in support of the minority rights. Their study considered a series of legal and

regulatory frameworks facing dividend policies in relation to the insiders‟ shareholdings. In the

same vein, Page, Jahera & Pugh (1996) considered the effect of anti-takeover amendment laws

on the dividend policy using 6 variables as proxies for dividend policy. It used a window of -/+

5days but failed to focus on the impact on the acquirer firms.

Several recent studies in M&A have been undertaken in assessing the shareholders wealth and

profitability of cross border and domestic mergers. Some studies have also examined target

and acquirers banks in various dimensions.

Hernando, Nieto & Wall (2008) examined the characteristics of domestic and cross border

targets and found that both have similar returns for their shareholders and also share common

features. Lensink & Maslennikova (2008) used 41 days window period to compare the

shareholders wealth incidental to the mergers of the cross border and domestic banks. Their

findings show that domestic deals create more value than the cross border deals. Infact, their

evidence shows that cross border mergers destroy shareholders value.

The above findings concur with Mangold et al (2008) who studied the shareholder wealth

effect of German banks and attest that the cross border mergers destroy shareholders wealth

while the domestic mergers create wealth. But the evidence are also contradicts some past

studies. Anand, Capron & Will (2005), for example, found cross border acquirers with

multinational scope are more wealth creative. The highlight of this finding is that, not only

cross border banks undertaking an acquisition create wealth but also banks with a multinational

profile. Further, Conn et al (2005) find that cross border acquirers have lower announcement

period and long run returns than the domestic acquirers. However, they add that while

domestic mergers yield negative shareholders returns, cross border mergers yield zero

abnormal returns and tend to remain negative after the mergers.

6

1.2 The Contribution of Thesis

The novelty of the present study lies in the methodology adopted in the event study and in its

examination of the effects of M&A and dividend policy. The study uses the total standardised

cumulative abnormal returns (CTSAR) as opposed to the traditional abnormal returns. The

uniqueness of the CTSAR is that it standardizes the abnormal returns (ARs) to cater for the

different degree of event impact. This is done by weighing the abnormal returns by the

standard deviation. The purpose of the standardization is to ensure that each abnormal return

has the same variance (Serra, 2002). Thus, by dividing each firm‟s abnormal residual by the

standard deviation over the estimation period, each residual has an estimated variance of 1.

The event window of the methodology employed in the present study is also extended. An

estimation period of 100 days and an event window of -/+ 30days are applied. Some studies

favour short windows of 1-5 days ( Andrade, Mitchell & Stafford, 2001; Bruner, 2005;

Mulherin & Boone, 2000; Campa & Hernando ,2004), which are prone to insensitivity to the

chosen model for the expected returns; whereas a longer window takes into consideration a

possible bid revision and competitions (Conn et al, 2005). It also ensures that all possible

factors that may influence the abnormal returns such as pre and post merger events and the

response of the market thereafter are captured. Aw & Chatterjee (2004) opined that a longer

window and thus estimation period ensures the sufficient observations for statistical accuracy

without running any risk of being far from the test period is achieved.

There are a handful of studies that have used long event windows in estimating the abnormal

returns in M&A. The table below provides a synopsis of some known M&A studies and the

event windows applied. Note that 60 days have are not commonly used in cross border versus

domestic studies in the EU.

7

Table 1.1 Some M&A Studies and the Length of the windows

Study

Domestic/

Cross border Country/Continent Period

Length of

window (days) -

/+

Aw & Chatterjee(2004) Cross border UK/Europe 1991-96 24

Campa & Hernando

(2006) Both Europe 1998-2002 30

Ang & Kohers (2001) Domestic USA 1984-96 36

Moeller et al. (2004) Domestic USA 1980-2001 36

Rad & Beek (1999) Both Europe 1989-96 40

Black et al (2003) Cross border USA 1985-95 60

Gregory & McCorrison

(2004)

Cross border Europe 1985-94 60

Goergen & Renneboog

(2004)

Both UK/Europe 1989-99 60

Baradwaj et al (1990) Both USA 1980-87 60

Lowinski, Schiereck &

Thomas (2004)

Both Swiss/Europe 1990-2001 63

The current study covers the period 1997 - 2007 and favours a long estimation period. These

years are significant because many EU mergers and acquisitions have occurred over the

periods. Moreover, the 11 years assessment period is necessary to properly capture any

economic events that could impact on the M&A results. Olson and Pagano (2000) argued that

the success or failure of the banks M&A should not be solely judged on the estimation and

announcement period but on the long term performance of the acquirers.

Although the CTSAR methodology is not new, no study has so far been undertaken in

conjunction with dividend policy. The current study estimates the abnormal returns of the

domestic and cross border bank acquirers and applies the CTSAR as a variable in the

dividend policy regression. The purpose is to determine the significance of the CTSAR in the

dividend policy formulation of the acquirer banks. The same technique is used in the

evaluation of the financial profitability of the acquirer banks where the CTSAR is used as a

proxy for shareholders wealth but correlated with the profitability variables.

8

In considering the dividend policy, two proxies are used: the dividend payout and the

dividend yield. The significance of both proxies is evaluated in the regressions. The study

therefore offers a unique triangulation involving the dividend payout, yield and the CTSAR

associated with bank M&A.

This research is very important as it makes a contribution to the ongoing M&A and dividend

policy puzzles. The findings are particularly useful for banks considering mergers as it

produces an empirical basis for restructuring dividend policy. The study lays further basis

for research into the both M&A and dividend policies as it attempts to capture a best fit

dividend model for banks.

Investors and shareholders would readily grasp and understand in concise terms, recent issues

relating to M&A and its dividend policy implications. As a result, they would be better

informed of the fundamental issues associated with M&A and its attendant impact on their

investments; whether in the domestic or foreign bank investments.

1.4 Research Objectives & Hypotheses

In summary, the study aims to assess the impact of bank Mergers and Acquisitions (M&A)

on the dividend policy of the cross border and domestic acquirer banks by investigating the

effect on their shareholders‟ wealth, financial performance and the dividend hypothesis.

The main research objectives are to:

1. Conduct empirical analysis using event windows and regression analysis to examine the

abnormal returns to shareholders of cross border merged banks relative to domestic

merged banks, based on a sample of 62 EU domestic and cross border bank mega M&A .

2. Review critically the literature on the performance effects of bank M&A and on the

dividend policy model for cross border and domestic acquirer banks.

3. Examine the theoretical linkage between M&A, shareholder return and dividend policy

of merged banks both for cross border and domestic acquirers.

9

4. Determine the factors affecting dividend policy and the most significant dividend policy

variables/measures for both the cross border and domestic acquirers.

The research is guided by a set of research questions and hypotheses which are formulated in

line with the research objectives. The following research questions are posed for the purpose

of guiding the research.

1) To what extent do the M&A announcement affect the shareholders returns of cross

border and domestic acquirers?

2) What effects do mergers and acquisitions have on the performance of cross border

and domestic bank acquirers?

3) What is the impact of mergers and acquisitions on the different dividend policy

measures of cross border and domestic bank acquirers and to what extent does this

geographical diversification affect the shareholders returns of the affected banks?

4) What other factors influence the dividend policy of cross border and domestic merged

banks?

Three research hypotheses which relate to the impact of M&A on the shareholders

wealth, performance and dividend policy of cross border and domestic acquirer banks,

are formulated and tested. One of the objectives of the present research is the

examination of the effect of M&A on abnormal returns. However, despite the

conflicting and mixed evidence in the literature concerning the impact of abnormal

returns on merged banks, the resulting impact is expected to be significant.

Therefore, a test of following hypotheses is to further examine the significance of the

M&A on shareholders returns, banks profitability and dividend policy. It is important

to note that the behaviour of abnormal returns reflects investor expectations about

profitability and dividends. Thus, the profitability and dividend regressions test the

accuracy of the investor expectations.

10

HI : The wealth effect and geographical diversification hypothesis: In contrast to domestic

acquisition, cross border mergers and acquisitions will not create more value.

Where diversification and possible reduction in cost is achievable, foreign acquisitions are

not expected to outperform domestic acquirers.

H2: The financial performance of cross border acquirers is significantly correlated with

M&A, and significantly higher than domestic acquirers.

Given their access to capital through foreign partners and investments, cross border bank

acquirers are expected to outperform the domestic acquirers.

H3. The dividend policy hypothesis: There is no significant difference in the shareholders

perception of dividend policy of cross border and domestic bank acquirer following a merger

or acquisitions.

It is expected that there would be a significant difference in the dividend policy, measured by

the dividend yield and payout, of both acquirers after the mergers and acquisitions.

1.4 The Organisation of the Study

The thesis is organised into eight chapters. The first chapter has provided a well coverage of

the background to the research, leading to the formulation of the objectives of the study;

including the research questions, hypotheses and justifications for the present study. The

next three chapters offer different strands of the broad literature that encompasses this stride.

Chapter 2 reviews the literature on dividend policy while chapter 3 discusses the various

theoretical debates and dividend hypotheses. Both chapters cover various forms and concepts

of dividend policy and review of the empirical studies. They also review the literature on

dividend policy as it relates to M&A. Chapter 4 reviews the literature on mergers and

acquisitions, abnormal returns and other contemporary issues associated with mergers and

acquisitions.

11

Chapter 5 covers the methodological framework of the study. It discusses the different

research techniques, data analysis and sampling issues and economic models applied in the

study. It describes the event windows methodology and other statistical tools such as

regression models, estimation methods and test statistics as used in the research.

Chapter 6 presents the first part of the data analyses and discussion of the results. It discusses

the research results based on the event window methodology. It also examines the domestic

and cross border M&A impact on bank financial performance. The first two hypotheses are

tested in this chapter. Chapter 7 provides further empirical analysis of the dividend policy

using both the dividend yield and payout. It uses hierarchical regression analysis to test the

dividend policy hypothesis. The effects exerted by the total standardized cumulative

abnormal returns (CTSAR) and other dividend policy variables are discussed and analysed in

the chapter.

Chapter 8 summarises the research findings, evaluates the results and offers

recommendations. The conclusions are drawn based on the findings of the study and are in

line with the objectives of the research. In addition to the recommendations, the chapter

makes suggestion for the furtherance of the research topic to a broader scope.

12

Chapter Two

Literature Review I

2.0 Introduction

The debate on dividend policy has remained an area of intense research in finance. Different

arguments and theories have been put forward to explain the different facets of dividend

policy. This chapter therefore explains the meaning and various interpretations of the

concept, factors determining dividend policy and theoretical models of dividend.

2.1 The Meaning and Types of Dividend Policy

In discussing the meaning of dividend policy, it is important to define a dividend. Various

definitions abound in the literature on the definition of dividend. A dividend is simply the

money that a company pays out to its shareholders from the profits it has made (Doughty,

2000). Such payments can be made in cash or by issuance of additional shares as in scrip

dividend. Davies & Pain (2002) however define it as the amount payable to shareholders

from profit or distributable reserves.

Companies that are listed in the stock exchange are usually obligated to pay out dividends on

a quarterly or semi annual basis. The semi annual or quarterly payment is referred to as the

interim dividend. The final payment, which is usually paid at the end of the financial year of

the company, is known as the final dividend. Dividends are normally paid after the corporate

tax has been deducted.

Dividend policy is primarily concerned with the decisions regarding dividend payout and

retention. Lease (2000) described it as the practice adopted by managers in making dividend

payout decisions. It details the amount of cash to be distributed to the shareholders and what

is to be retained by the bank for further investment. It is a decision that considers the amount

of profits to be retained and that to be distributed to the shareholders of the bank (Watson &

Head, 2004). The objective of a firm‟s dividend policy is to be consistent in the overall

objective of maximising shareholders wealth since it is the aim of every investor to get a

13

return from their investment. Economists, psychologists and the sociologists have all

attempted to explain investor behaviour in a number of ways and to relate the various

corporate dividend policies to the theories on the behaviour of individual investors.

Theoretically, there are different types of dividend policies. These include constant payout,

progressive policy, residual policy, zero policy and non- cash policy. Investors are seen to

belong to a particular group or clientele. This is because they tend to pitch their tent with a

particular policy that might suite them. This is referred as the clientele effect of dividend

policy (Hutchinson, 1995; Kolb & Rodriguez, 1996).

The amount of dividend to be paid out by firms could be influenced by the size of the firm.

Companies that are large in size are more than likely to pay dividend more often than small

firms. Larger firms also have higher agency costs and a relatively lower transaction cost than

the small firms. Dividend payout is inversely related to intrinsic business risk. Kalay (1980)

opined that companies with unstable earnings pay less in dividends in attempt to maintain a

stable dividend payout and to avoid the cost of borrowing from external sources.

2.2 Types of Dividend Policy

Constant or fixed policy: The Company pays out a fixed amount of its profit after tax as

dividend. Thus, the company maintains a fixed payout ratio of dividend. A company may as a

matter of policy, decide to constantly payout sixty percent of its after tax profit as dividend to

its shareholders and retain the remaining fraction. This type of policy allows the shareholders

the opportunity to clearly know the amount of dividend to expect from their investments in

the company. However as noted by Watson & Head (2004), the policy could be traumatic to

companies experiencing a volatile or fluctuating profit earning. This is because of the

uncertainty of its profit. If there are viable capital projects, the policy can be chaotic.

Progressive policy: Payments on dividend is on a steady increase usually in line with

inflation. This could result in increasing dividend in money terms. The firm uses the policy as

a ratchet. Every effort is made to sustain the increase even though marginal. Seldom, the

company may be constrained to cut down on dividend payout. This is to enable it sustain its

operations. This though is not a frequent action as it sends a wrong signal to investors. Firms

14

operating this policy will opt to avoid paying dividends during the period rather than

consistently cut down on the dividend (Kolb & Rodriguez, 1996).

Residual policy: Dividends are just what is left after the company determines the retained

profits required for future investment. This policy gives preference to its positive NPV (Net

Present Value) projects and paying out dividends if there are still left over funds available.

Dividend becomes a circumstantial payment paid only when the investment policy is

satisfied. There is a tendency therefore that this type of policy could give rise to a zero

dividend structure. Firms may need to modify this policy to ensure that investors of the

different clienteles are not chased out by a strict application of the policy (Kolb & Rodriguez,

1996).

Zero dividend policy: Some firms may decide not to pay dividend. This is especially

common in newly formed companies that rather require capital to execute its projects. All the

profit is thus retained for expansion of the business. Investors who prefer capital gains to

dividends because of taxation will naturally be lured by this kind of policy. This type of

policy is quite easy to operate and avoids all the costs associated with payment of dividends

(Watson & Head, 2004).

Alternative policies to paying cash: In order to give shareholders a choice between

dividends or new shares, the banks might choose to buy back shares. This is share or stock

repurchase. This has a significant advantage in terms of tax to the shareholder. While the

dividend is fully taxed just as ordinary income, the stock repurchase or buyback is not taxed

until the shares are sold and the shareholder makes a profit or capital gain (Ross, Westerfield

& Jordan, 2001). There is also the policy of stock dividends and splits. Shareholders are

given additional shares in lieu of cash as dividend (Brealey, Myers & Marcus, 1999).

2.3 Some Factors Determining Dividend Policy

The choice of a particular dividend policy by a bank is not usually accidental. It is tailored to

either meet the bank‟s need or the shareholders‟. Shareholders have different choice of

dividend depending on their needs. Firms also adopt policies that suite their peculiarity. Thus

some identifiable factors affecting dividend policy include but are not limited to the

following: reinvestment required for new capital projects, availability of other sources of

15

funds, existing cash levels and liquidity, market reaction to a change in dividend, tax,

shareholders preference for income or capital gain etc.

Market reaction to a change in dividend: Management often sticks to a pattern of dividend

payout. This can be explained by the signal conveyed by an alteration in dividend payment of

the company. A cut in dividends will give a negative signal to investors on the future

prospect of the company (Hutchinson, 1995). This might affect the share price of the

company. Thus, since management cannot afford to give such weak impression about its

company, it is likely to stick to its dividend policy that does not attempt to cut down on the

current payout.

Shareholders preference for income or capital gain: Investors are classified according to

their preferences. They are grouped into clienteles, „defined by their current income needs‟.

The wealthy investors may not have an immediate need for cash dividend. Others groups may

have strong need to receive cash as dividend. By targeting a particular clientele or group,

management may wish to tailor its dividend policy to suit the needs of its target investors

(Hutchinson, 1995).

There are reasons why an investor may wish to avoid or seek dividends other than cash. An

investor who is concerned about taxes will opt for little or no dividend to be paid to him. This

is because of the tax levied on dividend income. In attempt to avoid tax, some will prefer not

to receive dividend but to pay a capital gain tax at a later time when their shares are sold and

profit made. Thus, a firm will structure its policy to suit its investors (Kolb & Rodriguez,

1996).

Existing cash and liquidity level: The tendency to pay dividend is strictly based on

availability of funds. This is a residual dividend model, which holds that companies will

usually base their dividend policy on its cash flow and capital structure. Thus, where the

banks require capital for investment for a project, the retained earning is utilised and the

amount left for dividend will be affected. Only the residual earnings are used for the payment

of dividends (Samuels & Wilkes, 1987).

There are however, other issues that are crucial in the dividend policy of any business

organizations. These include the clientele effect, signalling hypothesis, dividend stability,

dividend extras, residual dividend model, share prices etc. Some investors prefer no

16

dividends while others may prefer large dividends. Investors will therefore invest in

companies that have suitable policy for them. A change in the amount of dividend sends a

signal to the investors. When dividend increases, it is a signal of growth and a reduction in

dividend signals otherwise. Thus, companies that have stabled or predictable dividend enjoys

the goodwill of investors.

Bank dividend policy is distinct and unique from other industries as it does not react to the

Rozeff model elements such as past growth rate, beta, and other acute factors (Casey and

Dicken, 2000). In a similar study, Baker, Powell and Veil (2001) segmented the factors

influencing dividend policy of financial and non-financial firms. Their findings include

pattern of past dividends, concern about maintaining a target capital structure, current degree

of financial leverage, shareholders need for dividend income, legal rules and constraints; such

as impairment of capital, the desire to send favourable signals to investors, the desire to

conform to the industry‟s dividend payout ratio, investment considerations such as the

availability of profitable investments, expected rate of return on firm‟s assets etc. Their

findings corroborate some of the factors under discussion, which depicts the congruence and

agreement of the dividend factors, irrespective of the industry.

Some firms have an established low dividend payout pattern even though their operating and

financial leverage increases. Dividend payout can thus serve as a quasi- fixed charge that can

replace the interest rate of borrowing or high fixed cost. Thus, as the fixed costs increases, the

dividend must reduce to avoid the additional cost of external borrowing. Moh‟d, Perry &

Rimbley (1995) argue that increases in dividend payout are often observed if a low

percentage of the firms‟ shares are held by the management of the company and and the

outsider shareholders become more dispersed.

In other words, higher dividend payouts are established when there is an increase in

institutional ownership of the firm. These views are in agreement with Rozeff (1982),

Easterbrook (1984) and Shleifer & Vishny, (2003) who argue that shareholders seek higher

dividend payout as they perceive the diminishing level of their control and that small

shareholders will want a high dividend payout as a compensation as well as attraction to

larger shareholders for their role in monitoring management‟s activities.

17

2.4 Some Theoretical Dividend Models

Corporate dividend policy has captured the interests of financial economists and has been

an issue of intensive theoretical models and empirical examinations over the last few

decades (Frankfurter & Wood, 1997). The controversy probably began with the

Modigliani & Miller (M&M)(1961) theory, which classified investors into dividend

clientele. It was later found in a later research that tax is responsible for marginal

alterations in the portfolio composition rather than the major differences predicted by

Miller & Scholes (1978). At the onset, when the M&M theory was postulated, most

financial practioners and many academics greeted the conclusion that tax had a marginal

effect on dividend policy with surprise (Baker, Powell & Veit, 2001).

Where dividends are taxed more heavily than capital gains, and where capital gains are not

taxed until realised, a bank that pays no dividends will be more attractive to taxable

individuals than a similar bank that pays dividends. This is because corporate investors are

taxed more providing a substantial incentive for investors to generate income by selling

their shares than to receive dividends.

Other models such as tax adjusted model which groups investors into tax dividend clienteles;

Masulis & Trueman (1988) Model of cash payments, Farrar & Sewlyn (1967) model of after

tax income of investors classifies investors into two clienteles. This is based on their

preference for dividend or capital gains. Auerbach‟s (1979) model of shareholder wealth and

Akerlof‟s (1970) signalling model, information asymmetric theories, Jensen (1986) free cash

flow hypothesis, Feldstein and Green (1983) theoretical dividend behavioural models and

Shefrin & Statman (1984) theory of self control are all measures to unmask firms‟ dividend

policy.

The strategy of tax sheltering of income especially by high tax investors was instrumental to

the Miller & Scholes (1978) and Miller (1986) attempt to inject rationality into the tax-

adjusted model. Scholes (1982) attributed any significant relationship between profit and

dividend to „dividend information effect‟ rather than the effect exerted by tax (Watson and

Head, 2004).

The shareholder wealth model assumes that the shareholders are the main priority in wealth

maximization rather the firm. Where there is an option of capital gain or tax on dividends, the

18

idea of wealth maximization seizes to be on the company‟s market values. Auerbach (1979)

in this model sees dividend as a child of circumstance; which results from a long-term under

valuation of capital of the firm and a reinvestment of the earnings after deducting the

expected returns of the firm and investors.

Akerlof (1970) came up with the signal model of corporate dividend policy. The crusaders of

the signal model affirm that when dividend is used as a method of presenting the message of

a firm‟s quality it bears a lower cost than other available alternatives. This implies that other

methods of putting across good performance of the firm are not reliable, thus the use of

dividend. This is possible because of the asymmetric nature of information regarding the

firm. The asymmetric of information between the owners of the business and the

management warrants the payment of dividends to restore dividend and adds value to the

firm‟s equity (Frankfurter & Lane, 1992). It is an aid to allay investors‟ fears and helps the

firm to remain competitive in the industry.

Other writers of the asymmetric information have developed thought on applying the agency

cost in modelling the dividend policy. This is exponential on the reduction of costs incurred

by management. It is firmly believed that huge dividend payment reduces the amount of

capital at the disposal of the managers. Thus, requiring management to seek fund in the

capital market to finance investment projects. This thus ensures that funds are effectively

monitored by the capital market and to ensure prudence in spending by management

(Easterbrook, 1984).

In line with this thought, Jensen (1986) postulated the free cash flow hypothesis, which is an

extension of the agency cost model and the signalling model. It hinges on the prudence of

managers on the use of funds available after financing the capital projects of the firm. The

remaining fund, he assumes, can lead to conflict of interest between the shareholders and the

management. Thus, the separation of the ownership and management affords the managers

the opportunity to demonstrate their worth. However, the funds left after dividend and

interest payments may become too marginal that unscrupulous managers have little to squash.

Feldstein and Green (1983) formulated a theoretical behavioural model of dividend in an

attempt to prove that dividend is a product of several factors. The model asserts that the tax

liability of dividends is marginal if compared with the transaction costs of trading shares

19

when the firm‟s profits are retained. Dividend payment reaffirms trust and confidence in

management and thus serves as a signal factor.

The theory of self-control is used by Shefrin and Statman (1984) to explain the investor‟s

preference for dividends to capital gains. The model posits that capital gains and dividends

are not close substitutes. Investors are less patient when dividend is delayed. This can be

verified when dividends are reduced. Thus dividend payout provides more checks on

spending level of management. Dividend provides opportunity to spend rather than save but

the opposite applies to the capital gains when earnings are retained.

The catering theory of dividend has become a front runner in the dividend model theories.

The principle behind the theory is that decisions to pay dividends are usually driven by

investors demand. Management therefore „cater‟ for investors by paying dividends to

shareholders who require it and not paying when the investors do not require dividends.

Baker and Wurgler (2004) argue that investors have uninformed and time varying demand for

dividend paying shares. This demand is not influenced by any arbitrage as the prices of the

payers and non-payers remain unperturbed.

Management would pay dividend when investors place higher prices on payers but avoid

payments if investors prefer nonpayers. The study used the catering dividend dynamics to

support that argument that managers cater for time varying investors in an attempt to

maximise share prices. Their results suggest that dividends are highly relevant to share values

but in different directions and times. Once dividends are initiated, any changes are influenced

by the level of profitability rather than the relative valuations of payer and nonpayers.

In a similar study of banks dividend payout, Casey and Dickens (2000) employed the Rozeff

(1982) model to examine dividend payout factors in which he found five variables to be very

significant in dividend payout policies of firms. The variables include beta, percentage of

insider ownership, past revenue growth rate, forecasted revenue growth and the number of

common stockholders. Those variables are equally useful in articulating dividend policies of

banks. Chang and Rhee (1990) added that financial leverage is a crucial factor in dividend

policy of firms. A firm that has a high financial leverage tends to have a high dividend payout

ratio. This however depends if the tax chargeable on the dividend income is higher than the

capital gains.

20

2.5 The Dividend Irrelevance Theory

The theory of the irrelevance of dividends was promulgated by Miller and Modigliani

(1961) in which they asserted that a firm does not let its dividend policy affect its

investment decisions, and if taxes and transaction costs are ignored, a firm‟s dividend

policy should have no impact on the value of its shares. This implies that the dividend a

bank pays does not affect the value of its shares or the returns to investors because the

higher the dividend paid, the less the investor receives in capital appreciation no matter

how the corporation‟s business decisions turn out.

Share valuation according to Miller and Modigliani is a function of the level of corporate

earnings which reflects a company‟s investment policy rather than a proportion of a

company‟s earnings paid out as dividends. The firm‟s choice of dividend policy given its

investment therefore is a choice of financing strategy. The basis of this proposition is

based on the assumption of a perfect capital market and no tax world; perfect only on the

assumption of a rational individual.

The importance of this argument is that investors will always prefer more wealth to less

and are indifferent as to whether a given increment to their wealth takes the form of cash

payment or an increase in the market value of their holdings. Perfect certainty implies

complete assurance on the part of every investor as to the future investment programme

and the future profits of the business. They propose that in a perfect capital market, no

buyer or seller of securities is large enough for their transaction to have an appreciable

impact on the ruling price.

However, there have been other scholars whose opinion and findings are in sharp contrast

with the irrelevance dividend theory. Gordon (1959) and Lintner (1964) have argued that

investors are very keen and happy with today‟s dividend as against a future gain that may

not be certain. The contention is that dividend policy is positively relevant to the valuation

of firms and shareholders will prefer a high dividend today to a highly uncertain capital

gain from a questionable future investment. This is often referred to as the „Bird in Hand‟

theory, since the gains from dividends are certain more than capital gains. If dividend is

21

preferred to capital gains then the dividend policy has an intrinsic role to play in the

valuation of a bank.

The behavioural life cycle propagated by Shefrin and Stateman (1984) which is based on

self control argues that investors would want to restrict themselves from consuming too

much in the present in order to enjoy future income. Investors do not want to dip their

hands into their capital and therefore allow themselves to consume current income such as

dividends. Investors therefore prefer to consume from dividends rather than capital gains

as such the issue of dividend is of relevance to investors.

Another dimension of the dividend puzzle is the proposition that dividend payout reduces

the market value of a company particularly whenever dividends are taxed more heavily

than capital gains. According to Brealey (1997), the fact that investors pay taxes at

different rates on their income relative to capital gains provide an incentive to hold

portfolios with different exposures to dividend yield even though this results in a less well-

diversified portfolio. Various studies have however refuted this argument that heavy

taxation on dividend is a strong incentive for individuals to prefer capital gains over

dividend distribution.

There is also the clientele effect of dividend policy. This effect makes management

reluctant to alter established payout ratios because of fear that such changes might incur

unwanted transaction costs to current shareholders. Shareholders are indifferent to whether

they receive dividends or capital gains (Watson and Head, 2004). Some types of

shareholders require dividends as a source of regular income and others solely prefer

capital gains or dividends due to the difference in tax treatment. Shareholders may prefer

firms to supply them with a pattern of dividend that fits their consumption pattern thereby

relieving them from the problem of having to change their cash flow especially where

pensioners are involved.

Thus, investors would prefer to consider in isolation their investments and invest in a way

that gives the highest opportunity to meet their particular goal with the lowest risk of

failure as defined by the investor. A significant change in a company‟s dividend policy

could give rise to unnecessary pressure on its shares.

22

2.6 Corporate Governance, Agency Theory and Dividend Policy Decision

The main idea behind the agency theory approach is that dividend payment reduces the

amount of cash held in the firm and consequently lowers the cost to investors. Miller and

Modigliani (1958) think of firms as collections of investment projects and the cash flows

these projects create, and hence naturally interpret securities such as debt and equity as

claims to these cash flows. Shleifer and Vishny (1997) found that the ownership structure,

along with a country‟s legal protection, is one of the most important determinants of

corporate governance.

In both the contractual framework of Jensen and Meckling (1976) and the residual control

rights framework of Grossman, Hart, and Moore (1986), the rights of investors are

protected and sometimes even specified by the legal system. In different jurisdictions,

rules protecting investors come from different sources such as company and security.

Other laws include bankruptcy and takeovers, competition, and also from stock exchange

regulations and accounting standards. It is argued by most writers that payment of

dividend is highly correlated with the type of governance structure.

In a study carried out by La Porta et al (2000), they argue that dividend policies vary across

legal regimes in ways consistent with a particular version of the agency theory of dividends.

Specifically, firms in common law countries where investor protection is typically better,

have higher dividend payouts than firms in civil law countries. Moreover, in common but not

civil law countries, high growth firms have lower dividend payouts than low growth firms.

Shareholder protection gives an investor the right to vote against a non performing

management in terms of mismanagement of the assets of the business. For example contract

law deals with privately negotiated arrangements, whereas company, bankruptcy, and

securities laws specifically describe some of the rights of corporate insiders and outsiders.

These laws and the quality of their enforcement by the regulators and courts are essential

elements of corporate governance and finance.

Jensen and Meckling (1976) define an agency relationship as a contract under which one or

more persons (the principal(s)) engage another person (the agent) to perform some service on

their behalf which involves delegating some decision making authority to the agent. Agency

23

costs arise with the separation of ownership and control of the firm because managers and

shareholders have different objectives. Agency cost is thus the sum of the monitoring of

expenditures by the owners, the bonding expenditures by the managers and a residual loss.

This residual loss results from the divergence between the agent decisions and those intended

by the principal that persists after the monitoring.

According to Jensen (1986), the conflicts of interest between shareholders and managers

about payout policies are especially severe when the firm generates substantial free cash

flow. The free cash flow is the cash flow in excess of that required to fund all projects that

have positive net present values when discounted at the relevant cost of capital. Therefore

with dividends, managers have fewer resources under control and the fact that capital markets

punish dividend cuts with large stock price reductions is consistent with the agency costs of

free cash flow.

Different shareholders have varying characteristics to influence the policies of the firm

especially if they have a role in management to limit the availability of funds to invest in

unprofitable projects. The investment policy of the firm based on this premises is therefore no

longer independent of its dividend policy. Conflicts of interest between corporate insiders,

such as managers and controlling shareholders on one hand, and outside investors such as

minority shareholders on the other hand, are central to the analysis of the modern corporation

(Cornett et al., 2008). Insiders can divert corporate assets to themselves through outright theft

dilution of outside investors through share issues, excessive salaries, asset sales to themselves

or other corporations they control at favourable prices, or transfer pricing with other entities

they control. Alternatively, insiders can use corporate assets to pursue investment strategies

that yield them personal benefits of control such as growth or diversification without

benefiting outside investors.

As pointed out by La Porta et al (1998), the extent of legal protection of outside investors

differs enormously across countries. Legal protection consists in both the content of the laws

and the quality of their enforcement. Shareholders, both insiders and outsiders through

corporate and other laws, are able to protect their investment against expropriation by

insiders. This right by shareholders cover a range of activities such as the right to vote on

important corporate matters and election of directors, the right to receive the same dividend

per share with insiders and the right to sue the company for default of contract. The idea is

24

that shareholders buy certain rights to the company once they invest although in most cases

these rights are effective when acted upon jointly.

In effect, common law countries appear to have the best legal protection of minority

shareholders, whereas civil law countries, and most especially the French civil law countries,

have the weakest protection. Evidence also suggests that good investor protection contributes

to the efficiency of resource allocation and to economic growth. Developing countries can as

such use shareholder protection as a means of attracting investors since an inflow of

investment opportunities lead to increased growth and high per capita income.

Dividend policy has been applied in an attempt to reduce agency cost. Rozeff (1982) found

that higher dividend payouts are constituted when a low proportion of the firm‟s equity is

held by insiders and non growth firms are found to pay greater dividend than firms in their

growth stage (Gaver & Gaver, 1993). The main idea behind the agency theory approach is

that dividend payments reduce the amount of cash held in the firm and consequently lowers

the cost to investors. By paying dividends, the insiders (management) are thus no longer in

position to use the corporate earnings to benefit themselves as such earnings are inversely

returned to the investors.

The theory of free cash flow seeks to expose the likely conflict of interest between the

shareholders and management. This is an integral aspect of the agency theory. Some frugal

managers who have the best interest of the shareholders are ordinarily expected to invest only

in profitable projects. Frankfurter & Wood (2002) alleged that the separation of the business

owners from the management of the firm offers the mangers the choice of making wrong

investments and possibly wasting funds.

The theory of free cash flow is predicated on the inefficient use of funds available to

managers after investing in projects with high positive net present values (NPV). The

assumption is that the funds remaining after the execution of profitable projects could cause

conflict of interest between the managers and shareholders. The free cash flow hypothesis is

predicated on the forgone findings that such funds are usually used to appease the

management interest rather than shareholders.

Another dimension of dividends is that it can be used as an upshot of legal protection of the

shareholders. La Porta et al. (2000) stated in their work that dividends serve as a substitute

25

for legal protection of shareholders in countries where shareholder protection is weak. Since

firms will have to come to capital markets to raise funds to finance their projects, they have to

create a reputation for good treatment of shareholders. Where there exist significant problems

between corporate insiders and outsiders, dividends can play a useful role since dividends are

better than retained earnings and the latter might never mature as dividends in the future.

This reputation is necessary in countries where shareholder protection is weak unlike in

countries with strong forms of protection.

Minority shareholders with the right to vote for directors under an effective system might use

their legal powers to force companies to pour-out cash, thus preventing insiders from using

too high a fraction of company‟s earnings to benefit themselves. Good investor protection

makes it possible for shareholders to vote for directors who offer better dividend policies by

selling shares to potential hostile raiders who then gain control over non dividend paying

companies. Insiders also get attracted to dividends since good investor protection make it

risky and costly to divert company assets

This assumption is however a contestable fact in countries with good shareholders protection

and where there are both good investment prospects and growth. If there are banks but with a

poor investment opportunities and growth, shareholders who feel unprotected will accept low

dividend payouts and high reinvestment rates from the banks with better investment

opportunities because they know that banks will have the potential of paying a higher

dividend. The shareholders of the companies with poor investment and growth will not feel

happy and protected under similar circumstances. Thus, with good investment and growth

potential, shareholders will feel protected to accept low dividend payout. This concurs with

La Porta et al‟s (2000) outcome agency model of dividends, which argues in favour of the

„bird in the bush‟ and not necessarily the „bird in hand‟ theory of dividends.

Other studies have attempted to examine the doctrine that dividends serve as a mechanism

for reducing agency costs and thus offer a rationale for the distribution of cash resources to

shareholders. Companies that are attempting to maximize shareholder wealth often seek to

reduce the agency problem by incurring additional costs, which helps to align the

managers to act in the interest of the outsider shareholders. Moh‟d, Perry & Rimbey

(1995) propose that the dividend mechanism provides an incentive for managers to reduce

the costs associated with the principal and agent relationship. Thus, distributing resources

26

in the form of cash dividends pushes managers to explore outsourcing in the capital

market, causing them to reduce agency costs as they subject themselves to the scrutiny of

the capital market.

Under this scenario, the optimum level of dividend payout is that which minimizes the

agency cost structure relative to the cost of raising the required capital. If the firm is forced

to raise external capital to replace the funds used in the payment of dividend, then

managers must reduce agency costs and reveal some new information as a prerequisite for

securing such loans. This argument which was initially proposed by Rozeff (1982), was

tested by Moh‟d et al (1995) by employing a cross sectional model of dividends which

regressed dividend payout against proxies for the agency cost and the transaction cost

trade off. Their result supports the position that an optimum dividend payout ratio can

exist even without any consideration for tax implications.

In a similar vein, Easterbrook (1984) argues that outside shareholders are active in seeking

to draw capital from the company to ensure that managers are subjected to examination by

the capital market as a way of measuring up with the requirement of loans. The payment

of dividend is thus considered to be a measure of causing the firm to undergo a cross

examination by the capital market or third party audit in order to secure such needed

capital. The implication of such is that the managers are forced into revealing relevant

information and reduce agency costs in order to secure such loans. Thus, shareholders are

happy to bear the cost implications of such new funding to realize the greater benefits

associated with agency cost and information asymmetries.

2.7 Taxation and Dividend Policy Theories

The relationship between taxes and dividend policy has attracted much academic interest.

Financial theorists such as Brennan (1970), Masulis & Trueman (1988) have stipulated that

taxes affect organizational corporate dividend policy. Under this scenario, changes in

corporate dividend payout would be expected whenever the government changes its income

tax policy (Wu, 1996). However, this does not always apply especially in the banking

27

business. Linter (1956) had asserted that the major determinants of dividend policy are the

anticipated level of future earnings and the pattern of past dividend.

This discrepancy may have underpinned Modigliani & Miller‟s (1961) theory, which

provided a platform for the debate and research on dividend policy. The banking sector is of

interest in this research because of the structure of its dividend policy as will be revealed in

the study.

Dividends are usually paid to owners or shareholders of business at specific periods. This is

apparently based on the declared earning of the company and the recommendations made by

its directors. Thus, if there are no profits made, dividends are not declared. But when profits

are made, the company is obligated to pay corporate tax including other statutory taxes to the

government. This is an essential corporate responsibility particularly with profit making

companies. The taxes no doubt reduce the profits available at the disposal of the

organizations, either to be retained or distributed as a dividend to shareholders of the

company.

For decades, several postulations and assumptions have been made regarding whether such

taxes paid by organizations actually affect their pattern of dividend policy. Dividend policy is

the trade-off between retaining earnings and paying out cash or issuing new shares to

shareholders. Some firms may have low dividend payout because management is optimistic

about the firm‟s future and therefore wishes to retain their earnings for further expansion. It is

hard to deny that taxes are important to investors. Although, dividend affects the shareholders

tax liability, it does not in general alter the taxes that must be paid regardless of whether the

company distributes or retains its profit (Brealey, Myers & Marcus, 1999, and Nnadi &

Akpomi, 2008).

The tax adjusted model2 groups investors into tax dividend clienteles. It purports that

investors are happy with shares that offer higher returns but will accept lower value for shares

due to the tax liability of the dividend. However, because investors have different tax

liabilities, their choice of investment based on company‟s dividend policy will be at variance

with each other. This agrees with the propositions of Modigliani (1982) and Masulis and

Trueman (1988) models of grouping investors into tax clienteles. It is believed that dividend

2 The dividend policy hypotheses, including the tax adjusted model are fully discussed in the next chapter.

28

payment decreases or increases in opposite direction with tax liability (Frankfurter & Wood,

1997).

The strategy of tax sheltering of income especially by high tax investors was instrumental to

the Miller & Scholes (1978) and Miller (1986) attempt to inject rationality into the tax-

adjusted model. An investor can choose to either buy or decline to buy shares with high

dividend as a technique of avoiding the apparent tax liability on such shares. Investors could

turn to tax free shares to counter balance it with the tax incidental on dividend paying shares.

However, DeAngelo & Masulis (1980) disagree on the effective effect of the tax sheltering

strategy but rather advanced personal tax shelters to minimize or totally avoid tax. The study

reaffirms Bradley, Capozza and Seguin (1998) hypothesis on dividend policy and cash flow

uncertainty. They assert that when a shareholder is confronted with a greater tax rate on

dividend rather than capital gain, then the coefficient associated with dividends will be

significantly affected.

However, the finding of Black and Scholes (1974) which sets out to examine the impact of

profit on the dividend policy of companies fell short of any concrete findings as no

correlation was established in the two variables. This contrasted with a later finding of

Ramaswamy (1979) in which a significant relationship was established. But Miller and

Scholes (1982) attributed any significant relationship between profit and dividend to dividend

information effect‟ rather than the effect exerted by tax (Watson and Head, 2004). .

The after tax income model of investors also classifies investors into two clienteles. This is

based on their preference for dividend or capital gains. Farrar & Selwyn (1967) argue that

share repurchase rather than dividend should be implored in distributing the company‟s

profits to shareholders. This model was further investigated by Brennan (1970) into an

equilibrium framework. But as stated by Frankfurter & Wood (1997), equilibrium with

dividend paying firms is not consistent with a zero required return per unit of dividend yield.

Financial theorists such as Wu (1996) opined that evasion of taxes by companies is a key

factor in the determination of the extent to which its dividend policy is affected by tax. Miller

& Scholes (1982) however admit that taxes weigh tremendous influence on corporate

dividend structure. In turn, they contend the effects of dividend policy on company‟s share

price. What implications taxes have on dividend policy is a matter of intense debate. The

29

dividend irrelevant theory of Modigliani & Miller (1961), which assumes a perfect market, is

still very much held in contention but its principles underlines most dividend policies.

Three contemporary schools of thought have emerged with theories all attempting to explain

the dividend structure and the impact of taxation. The tax preference theorists agree that an

increase in dividend payout reduces value. They reason that since dividends are taxed as

ordinary income and capital gains, investors will prefer capital gains to dividends. These

groups strongly assert that whenever dividends are taxed more heavily than capital gains, the

firms would pay the lowest cash dividend they can get away with. The available cash would

be retained and reinvested or used to repurchase the shares.

The dividend irrelevance theory assumes the absence of brokerages or taxes. This may be

seen as presumptuous and unrealistic in practice, as taxes are inevitable in the business world.

The seemingly popular theory led by Modigliani & Miller (M&M) claims that dividend

policy makes no difference in a world without taxes, transaction costs, or other market

imperfection. They further postulated that companies can raise shares whenever they need

capital to finance the business.

Several studies aimed at correlating tax and dividends have been conducted with inconclusive

results. Miller and Modigliani‟s (M&M) theory assumes that taxes do not affect dividend

policy of the firm in a perfect market. Thus, investors‟ decisions are independent of dividend

considerations. However, when the tax on dividend is higher than that of the capital gain of

the investor, the issue of dividend policy of the firm becomes a concern to the investors. Chu

(1996) studied the effect of tax on personal income and posits that dividend tends to expose

investors to more tax liability than capital gains. Therefore, they will opt for lower dividend

to minimize the tax liability on investment. The findings also shows that the after tax income

of shareholders is greater when the firm pays no dividends, where the capital gains is lower

than the tax consequential on the dividend.

In the US, several studies have been undertaken to examine the effect of the various tax

reforms on corporate dividend policy. Bolster & Janjigian (1991) researched on the 1986 tax

laws in US but found no concrete evidence of any correlation of any increase in dividend

following the reform. However, Papaioannou and Savarese (1994) found an increase in the

dividend payout for the average firms after such reforms. This is attributed to the reduction in

30

income tax injected in the tax system and thus aligns with the preposition that tax has a

tremendous effect on the dividend policy of firms.

One identifiable problem of inferring the tax effect on dividend is the variation in tax rates

faced by different investors. For investors whose personal tax liabilities are less than their

capital gain tax liability (which underscores their interest in dividend rather than capital

gains), the tax effect does not apply (Hutchinson, 1995). Different tax situations would ignite

diverse effects on the share price of the firm. The rise in dividend can inversely be attributed

to the tax provisions on such dividends.

2.8 The dividend Signalling hypothesis

A major factor in the dividend payout strategy is the belief that the payout has a potential

effect of passing information to the public. Presumably, managers possess vital information

about the organisation that may not be in the domain of the public. Such information gap

between the insiders and outsiders could cause an inaccurate valuation of the firm and thus

renders the share price as an inappropriate measure of the firm‟s value.

In order to close this information gap, managers might opt to disclose the information to the

public to paint a measurable value of their organizations. Baskin and Miranti (1997) assert

that dividend can play the role of providing a veritable tool in conveying the private

information they hold on the organization to the market. Since investors use cash flows to

equity as measure of valuing a firm, dividend might give implicit information about the

banks‟ prospect.

The argument surrounding the dividend hypothesis was postulated by M&M (1961) in which

they argue that when markets are imperfect, the share prices may respond to changes in

dividends. This implies that dividend announcements may be an indication of the banks‟

future earnings. Investors can draw inferences from the information conveyed by the

dividend announcements. This signal can be detected by the sustainability or the changes in

the pattern of the dividends. This assumption is commonly referred to as the signalling

hypothesis.

31

For the signalling hypothesis or information content of dividends to be realistic, two

conditions are very vital. First, the bank managers must possess private information about the

future potentials of the organization which they are happy to convey to the market. Second,

the signal conveyed must be correct. False or wrong information about poor prospect of the

banks must not be falsified to deceive the public by increasing dividends. Koch and Shenoy

(1999) emphasized the importance of the public trust and the reliance on insider information

in making decision about their investments.

Thus an increase in dividend may be seen as signal of profitability by investors and could

increase the share price of the firm. A cut in dividend would be perceived as a bad signal of

the firm which can adversely affect the share price. This perception concurs with Lintner

(1956) model which suggests that firms increase dividend when there is a steady increase in

earnings. Thus dividend increase signals a long term profitability of the firm. This

assumption is consistent with the dividend smoothing theory which asserts that firms will

smooth dividends over time and not make substantial increase in dividend payout except

where such increase is sustainable in the future ( Lipson et al, 1994).

2.9 Summary of Literature

The chapter has examined some debates in the finance literature supporting various concepts

of dividend policy and their implications. The dividend irrelevance argument led by

Modigliani & Miller (1961) has also been discussed; though this will be further debates on

the theory will be reviewed in the next chapter.

Corporate governance, agency theory and its impact on the dividend decisions have also been

examined. Various studies and the theories underpinning their consideration by firms in their

dividend policy are considered important variables, for instance, Rozeff (1980), La Porta et

(2000). The implications of taxation in dividend as well as the signalling effect of dividend

have been considered.

32

Chapter Three

Theoretical Debates and Dividend Policy Hypotheses

3.0 Introduction

Organisations are generally faced with two critical operational decisions: investment and

financing decisions. The investment or capital budgeting decision is concerned with what

assets the firm should acquire while the financing decision is concerned with how these assets

should be financed. However, when the firm begins to generate profits, there may a be need

to consider whether the firm should distribute all or a proportion of earned profits in the form

of dividends to the shareholders, or whether it should be ploughed back into the business.

Presumably, in taking any course of action, managers should concentrate on how to maximise

the wealth of shareholders for whom the firm is being managed. Managers must not only

consider the question of how much of the company‟s earnings are needed for investment, but

also take into consideration the possible effect of their decisions on share prices (Bishop et

al., 2000).

The term „dividend policy‟ refers to “the practice that management follows in making

dividend payout decisions or, in other words, the size and pattern of cash distributions over

time to shareholders” (Lease et al., 2000, p.29). The concept of dividend policy has engaged

managers since the birth of the modern commercial corporation and has yet remained among

the most contested issues in finance. The study of dividend policy has captured the attention

of finance scholars since the middle of the last century. They have attempted to solve several

issues pertaining to dividends and formulate theories and models to explain corporate

dividend behaviour.

The dividend enigma has not only been an enduring issue in finance, it also remains

unresolved. Over thirty years since Black (1976) described it as a “puzzle”, an enormous

amount of research has been undertaken in an attempt to unravel the puzzle. Allen, Bernardo

and Welch (2000) summarised the prevailing consensus view when they concluded that

despite the number of theories that have been put forward in the literature to explain their

pervasive presence, the issue of dividends remains one of the thorniest puzzles in corporate

33

finance. This chapter can only give a broad overview of the main issues, debates and

hypothesis relating to dividend policy.

3.1 Brief History of Corporate Dividend Policy

The issue of corporate dividends has a long history and, as Frankfurter and Wood (1997)

observed, is bound up with the development of the corporate status. Corporate dividends date

back at least to the early 16th century in Holland and Great Britain when the captains of

sixteenth century sailing ships started selling financial claims to investors, which entitled

them to share in the proceeds of the voyages3. At the end of each voyage, the profits and the

capital were distributed to investors, liquidating and ending the venture‟s life. By the end of

the 16th century, these financial claims began to be traded on open markets in Amsterdam

and were gradually replaced by shares of ownership. It is worth mentioning that even then

many investors would buy shares from more than one captain to diversify the risk associated

with this type of business.

At the end of each voyage, the enterprise liquidation of the venture ensured a distribution of

the profits to owners and helped to reduce the possibilities of fraudulent practice by captains

(Baskin, 1988). However, as the profitability of these ventures was established and became

more regular, the process of liquidation of the assets at the conclusion of each voyage became

increasingly inconvenient and costly. The successes of the ventures increased their credibility

and shareholders became more confident in their management (captains), this was

accomplished by, among other things, the payment of “generous dividends” (Baskin, 1988).

As a result, these companies began trading as going concern entities, and distributing only the

profits rather than the entire invested capital. The emergence of firms as a “going concern”

initiated the fundamental practice of firms to decide what proportion of the firms‟ income

(rather than assets) to return to investors and produced the first dividend payment regulations

(Frankfurter and Wood, 1997). Gradually, corporate charters began to restrict the payments

of dividends to the profits only.

3 This type of business was called Commenda. Under the commenda, the commendator provided the

capital and the commendatarius managed the investment (Walker, 1931, p.97).

34

The ownership structure of shipping firms gradually evolved into joint stock companies. But

it was chartered trading firms that adopted the joint stock status. By 1613, the British East

India Company issued its first joint stock shares with a nominal value without any distinction

made, between capital and profit of the firm (Walker, 1931). In the 17th century, the success

of this type of trading company attracted others and thus spread to include other activities

such as mining, banking, clothing, and utilities. Indeed, in the early 1700‟s, excitement about

the possibilities of expanded trade and the corporate form saw a speculative bubble form,

which collapsed spectacularly when the South Sea Company went into bankruptcy. The

Bubble Act of 1720 effectively slowed, but did not stop, the development of the corporate

form in Britain for almost a century (Walker, 1931).

In the early stages of corporate history, managers realized the importance of high and stable

dividend payments. In some ways, this was due to the analogy investors made with the other

form of financial security then traded, namely government bonds. Bonds paid a regular and

stable interest payment, and corporate managers found that investors preferred shares that pay

regular dividends. For example, Bank of North America in 1781 paid dividends after only six

months of operation, and the bank charter entitled the board of directors to distribute

dividends regularly out of profits. Paying consistent dividends remained one of paramount

importance to managers during the first half of the 19th century (Frankfurter & Wood, 1997).

In addition to the importance placed by investors on dividend stability, another issue of

modern corporate dividend policy to emerge early in the 19th century was that dividends

came to be seen as an important form of information. The scarcity and unreliability of

financial data often resulted in investors making their assessments of corporations through

their dividend payments rather than reported earnings. In short, investors were often faced

with inaccurate information about the performance of a firm, and used dividend policy as a

way of gauging what management‟s views about future performance might be. Consequently,

an increase in divided payments tended to be reflected in rising stock prices.

As corporations became aware of this phenomenon, it raised the possibility that managers of

companies could use dividends to signal strong earnings prospects and to support a

company‟s share price because investors may read dividend announcements as an indication

of growth in the firm‟s earnings. Frankfurter and Wood (1997) describe how in the 1920‟s

35

and 1930‟s corporate managers increasingly responded to the recognition that investor

behaviour is affected by dividend policy. Managers thus used dividend policy to affect share

prices through investor reactions to payout ratios and to signal expected future earnings.

Thus in summary, the development of dividend payments to shareholders has been tied up

with the development of the corporate entity. Corporate managers realized early the

importance of dividend payments in satisfying shareholders expectations. They often

smoothed dividends over time believing that dividend reductions might have unfavourable

effects on share price and therefore, used dividends as a device to signal information to the

market. Moreover, dividend policy is believed to have an impact on share price. Since the

1950‟s, the effect of dividend policy on firm value and other issues of corporate dividend

policy have been subjected to a great debate among finance scholars. The following section

discusses the various theoretical and an empirical arguments of dividend hypotheses.

3.2 Dividend Policy Theories

The previous section has established that dividend policy was bound up with the development

of the corporate entity. It was seen that the emergence of dividend policy as important to

investors was, to some extent, driven by the evolving state of financial markets. Investing in

shares was initially seen as analogous to bonds, so regularity of payments was important. It

was also seen that in the absence of regular and accurate corporate reporting, dividends were

often preferred to reinvested earnings, and often even regarded as a better indication of

corporate performance than published earnings accounts. However, as financial markets

developed and became more efficient, it was thought by some that dividend policy would

become increasingly irrelevant to investors. Why dividend policy should remain so evidently

important has been theoretically controversial.

Three main contradictory theories of dividends can be identified. Some argue that increasing

dividend payments increases a firm‟s value. Another view claims that high dividend payouts

have the opposite effect on a firm‟s value; that is, it reduces firm value. The third theoretical

approach asserts that dividends should be irrelevant and all effort spent on the dividend

decision is wasted. These views are embodied in three inter-related theories of dividend

policy: high dividends increase share value theory (or the so-called „bird-in-the- hand‟

36

argument), low dividends increase share value theory (the tax-preference argument), and the

dividend irrelevance hypothesis. Dividend debate is not limited to these three approaches,

however, several other theories of dividend policy have been presented, which only serve to

increase the complexity of the dividend puzzle. Some of the more popular of these arguments

include the information content of dividends (signalling), the clientele effects, and the agency

cost. These issues are discussed later in this chapter following the three main theories.

3.3 Dividend Irrelevance Hypothesis (DIH)

3.3.1 Background & Implications

Before the emergency of the Miller and Modigliani (1961) theory (M&M, hereafter) on

dividend policy, a common belief was that higher dividends increase a firm‟s value. This

belief was mainly based on the so-called “bird-in-the-hand” argument, discussed in more

detail in the next section. Graham and Dodd (1934), for instance, argued that “the sole

purpose for the existence of the corporation is to pay dividends”, and firms that pay higher

dividends must sell their shares at higher prices (cited in Frankfurter et al., 2002 p201).

However, as part of a new wave of finance in the 1960‟s, M&M demonstrated that under

certain assumptions about perfect capital markets, dividend policy would be irrelevant.

Given that in a perfect market, dividend policy has no effect on either the price of a firm‟s

stock or its cost of capital, shareholders wealth is not affected by the dividend decision and

therefore shareholders would be indifferent between the choice of dividends and capital

gains. The reason for this indifference is that shareholder wealth is affected by the income

generated by the investment decisions a firm makes, not by how it distributes that income.

Therefore, in M&M‟s world, dividends are irrelevant. M&M argued that regardless of how

the firm distributes its income, its value is determined by its basic earning power and its

investment decisions. They opined that given a firm‟s investment policy, the dividend payout

policy it chooses to follow will affect neither the current price of its shares nor the total

returns to shareholders.

This implies that investors calculate the value of companies based on the capitalised value of

their future earnings which is not affected by whether firms pay dividends or not and how the

37

firms set their dividend policies. M&M go further and suggests that, to an investor, all

dividend policies are effectively the same since investors can create “homemade” dividends

by adjusting their portfolios in a way that matches their preferences.

M&M based their argument upon idealistic assumptions of a perfect capital market and

rational investors. The underlying assumptions of a perfect capital market for the dividend

irrelevancy hypothesis to hold can be summarized as follows: (1) no differences between

taxes on dividends and capital gains; (2) no transaction and flotation costs incurred when

securities are traded; (3) all market participants have free and equal access to the same

information (symmetrical and costless information); (4) no conflicts of interests between

managers and security holders (i.e. no agency problem); and (5) all participants in the market

are price takers.

Miller and Modigliani (1961) point out that if the firm does not let its dividend policy

affect its investment decisions, and in absence of taxes and transaction cost, a firm‟s

dividend policy should have no impact on the value of its shares. The dividend a company

pays do not affect the value of its shares or the returns to investors because the higher the

dividend paid, the less the investor receives in capital appreciation no matter how the

corporation‟s business decision turn out (Black 1976 ). The firm‟s valuation follows the

principle that the price of each share must be such that the rate of return (dividends plus

capital gains per dollar invested) on every share will be the same throughout the market

over any given interval of time otherwise arbitrage will set in (Miller and Modigliani

1961).

Share valuation according to Miller and Modigliani is a function of the level of corporate

earnings which reflects a company‟s investment policy rather than a proportion of a

company‟s earnings paid out as dividends. The firm‟s choice of dividend policy given its

investment therefore is a choice of financing strategy. The basis for this proposition is

based on the assumption of a perfect capital market, no tax world, and perfect certainty

and on the assumption of a rational individual.

Investors always prefer more wealth to less and are indifferent as to whether a given

increment to their wealth takes the form of cash payment or an increase in the market

value of their holdings. Perfect certainty implies complete assurance on the part of every

38

investor as to the future investment programme and the future profits of every corporation.

In a perfect capital market, no buyer or seller of securities is large enough for his

transaction to have an appreciable impact on the then ruling price.

However, some studies have found that dividends are not irrelevant in the way Miller and

Modigliani proposed. Prior to M&M, Gordon (1959) suggested that dividend policy is

positively relevant to the valuation of firms and argues that shareholders prefer a high

dividend today to a highly uncertain capital gain from a questionable future investment.

This is known as the “Bird in Hand” fallacy, since the gains from dividends are certain BY

more than capital gains. If dividend is preferred to capital gains then the dividend policy

has an intrinsic role to play in the valuation of a company.

Another proposition on the issue of dividend policy points out that, an increase in the

dividend payout reduces the market value of a company whenever dividends are taxed

more heavily than capital gains. According to Brealey (1997), the fact that investors pay

taxes at different rates on their income relative to capital gains provide an incentive to hold

portfolios with different exposures to dividend yield even though this results in a less well-

diversified portfolio. Various studies have, however, refuted this argument that heavy

taxation on dividend is a strong incentive for individuals to prefer capital gains rather than

dividend income (Nnadi & Akpomi, 2008, Poterba, 2004, Poterba & Summers, 1984).

3.3.2 Proof of Irrelevancy and Empirical Evidence

The basic valuation model of common stock, that is, the dividend discount model (DDM)

would be a necessary step towards understanding the M&M proposition of dividend

irrelevancy. Generally, the DDM states that the value of a stock is a function of future

dividends (as a proxy for earnings) and the required rate of return on the stock. For example,

the value of a share at time zero (today) is simply the present value of all future dividends

discounted at an appropriate discount rate. This can be expressed as follows:

39

…………………………………………………………………………………………………….1

Where, P0 is the current share price; t is the time of the dividend; Dt is the dividends paid at

period t; and rt is the required rate of return for period t. The DDM suggests that future

discounted dividends (Dt) are the underlying determinant of the value of the current share

price (P0), and not any future share price. The share price is the critical determinant of the

firm value (V0). Accordingly, more dividends increase the value of the firm, other things

being equal. This intuition was generally accepted by most of the economists until M&M

published their paper, initiating a new direction of research in the dividend policy literature.

In a perfect capital market the required rate of return for an investor on equity shares (r) is

equal to dividends plus capital gains. This assumes one period world. For simplicity of the

exposition,

r = D1 + (P1 +P0) …………………………………………………2

P0

where, P0 is the current market price of shares; P1 is the expected market price at the end of

period one (the ex-dividend price of the share); D1 is the dividend at the end of the period.

The above equation can be re-arranged to give the current market price of shares as:

P0 = DI + P1 …………………………………………………………..3

(1+r)

Note that the above equation can be derived also from the basic valuation model (eq.1

above). Now, if we let n be the number of shares outstanding at time zero, then the current

value of the firm (Vo) is:

40

nP0 = V0 = nD1 + nP1 ……………………………………………………………………….4

(1+r)

Recall that M&M stated that in a perfect capital market, firm value is independent of

dividend policy. To illustrate, we can employ the sources and uses of funds equation. Given

the assumption that the market value of the firm is independent of capital structure

(Modigliani and Miller, 1958), debt financing is excluded from the analysis. On one hand, the

firm‟s sources of funds are cash flow from operations (CF1) and any new equity financing

(mP1), where m is number of shares issued at time one. On the other hand the uses of funds

are dividends payments (nD1) and investment made during the period (I1). Since sources must

equal the uses of funds, thus:

CF1 + mP1 = nD1 +I1 …………………………………………………………………5

Rearranging the above equation, we obtain

nD1 = CF1 + mP1 –I1 ……………………………………………………………………6

Substituting equation (6) into equation (4), for D1 gives:

VO = CF1 +nP1 –I1 +nP1 ………………………………………………….7

(1 +r)

VO = CF1 – I1 + (n +m) P1 ………………………………….8

(1 + r)

Since, (n+m) P1 = V1 hence;

V0 = CF1 –I1 + V1 .........................................................................................9

(1+r)

As dividends do not appear in the equation (9), and since operating cash flows (CF1),

investments (I1) and required rate of return (r) are not functions of dividend policy (either by

41

their nature or by assumption), the value of the firm is thus independent of its current

dividend policy (M&M, 1961). This analysis can be repeated for more periods and the results

will remain the same; i.e., the value of the firm is unaffected by dividend policy. The notion

that in perfect capital markets dividend policy should be irrelevant is a logical extension of

the neoclassical proposition of perfect competition into financial economics. Its elegance and

simplicity were recognised by M&M.

The above discussion suggests that the firm‟s investment policy is the key determinant of its

value and dividend policy is the residual. Operating cash flows depend on investments. In

other words, the firm‟s investments in positive net present value (NPV) projects will increase

the cash flows from operation, which is the only way to increase the value of the firm. In

summary, given the assumptions of perfect capital markets, the firm‟s future cash flow from

investment activities is the sole determinant of the value of the firm. The firm‟s payout policy

must therefore be independent of its value (Bishop et al., 2000).

The M&M dividend irrelevance proposition has provided the foundation for much

subsequent research on dividend policy. However, as argued by Ball et al. (1979), the

empirical tests of M&M‟s dividend irrelevance theorem have proven difficult to design and

to conduct. Recall that M&M built their conclusions on a certain set of assumptions of perfect

capital markets. Relaxing one or more of these assumptions has formed the basis for most of

theoretical and empirical studies.

In line with the dividend irrelevance hypothesis, Black and Scholes (1974) examined the

relationship between dividend yield and stock returns in order to identify the effect of

dividend policy on stock prices4. They constructed 25 portfolios of common stocks listed on

the New York Stock Exchange (NYSE), extending the capital asset pricing model (CAPM)5

to test the long run estimate of dividend yield effects. The study employed the following

regression model:

4 It is worth pointing out that Black and Scholes‟ study tested the tax-effect hypothesis, but it is presented here

because its conclusion strongly supports the M&M‟s irrelevance proposition. 5 The CAPM was developed alongside M&M‟s work on efficient markets and forms an integral part of the

framework of modern finance theory. The CAPM is expressed as E(Ŕi) =Rf + { E(Ŕm) - Rf} βi; where E(Ŕi) is

the expected return for security I, Rf is the risk free interest rate, E(Ŕm) is the expected return on the market

portfolio and βi is the beta for security i.

42

E(Ŕi) = γ0 + {E (Ŕm) – γ0}βi + γ1 (∂I -∂m) + εi ………………………………………………………10

∂m

where, E(Ŕi) is the expected return on portfolio i, E (Ŕm) is the expected return on the market

portfolio, γ0 is an intercept to be compared with short-term risk free rate R, βi is the

systematic risk of portfolio i, γ1 is the impact of dividend policy, δ1 is the dividend yield on

portfolio i, δm is the dividend yield on the market, and εi is the error term.

Black and Scholes used a long-term definition of dividend yield (previous year‟s dividends

divided by the year-end share price). Their results showed that the dividend yield coefficient

(γ1) is not significantly different from zero either for the entire period (1936-1966) or for any

of shorter sub-periods. This implies that the expected returns either on high or low yield

stocks are the same. They were therefore unable to prove that differences in yield lead to

differences in stock prices. That means that neither high-yield nor low-yield payout policy of

firms seemed to influence stock prices. Their conclusion lent important empirical support to

M&M‟s dividend irrelevance argument. Other studies by leading financial economic

researchers such as Miller and Scholes (1978, 1982), Hess (1981) Miller (1986), and

Bernstein (1996) provided evidence in support of the dividend irrelevance hypothesis (DIH).

While some empirical research supported the DIH, other are unsupportive or provided

evidence directly challenging the irrelevance hypothesis6. Building on Black and Scholes‟s

work, Ball et al. (1979) examined the effect of dividends on firm value using Australian data

over the period 1960 to 1969. They however, failed to find conclusive evidence to support

M&M‟s irrelevance proposition. Baker, Farrelly and Edelman (1985) surveyed the chief

financial officers (CFOs) of 562 firms listed on the New York Stock Exchange (NYSE) from

three industry groups (150 utilities, 309 manufacturing, and 103 wholesale/retail). Based on

318 responses, they found that respondents strongly agreed that dividend policy affects

common stock prices.

6 A representative sample of that debate would include: Lintner (1962), Gordon (1963) Walter (1963), Baumol

(1963), Brigham and Gordon (1968), and Van Horn and McDonald (1971). See e.g. Baker, Veit, and Powell

(2001) and Baker, Powell and Veit (2002a, 2002b).

43

In another survey study, Partington (1985) found that Australian senior managers viewed

dividend payments as a way to satisfy shareholders and support the share price. Baker and

Powell (1999) surveyed 603 CFOs of US firms listed on the NYSE, and observed that 90% of

respondents believed that dividend policy affects a firm‟s value as well as its cost of capital.

Further studies by the same authors tend to confirm that dividend policy actually matters in

the determination of firm value. Other studies including Siddiqi (1995) and Casey and

Dickens (2000) have provided evidence inconsistent with dividend irrelevance hypothesis.

Little evidence on the M&M dividend irrelevance hypothesis exists for emerging markets.

Ben, Naceur and Goaied (2002) used unbalanced panel data to estimate a random effects

Probit model to test whether the probability of creating future value of the firms relates to

dividend policy, financial policy, and profitability. The dividend policy (measured by payout

ratio) and financial structure (measured by debt to total assets) were found to be insignificant.

They concluded that their evidence supported the M&M irrelevance propositions of dividend

hypothesis. In contrast, Omet and Abu-Ruman (2003) provided evidence inconsistent with

the DIH. Omet and Abu-Ruman surveyed the CFOs of 47 manufacturing companies and

identify their views about dividend policy. Based on 33 responses, the researchers observed

that most of the Chief Finance Officers ( CFOs) questioned strongly agreed that dividend

policy affects share prices. However, given the small sample size in both the Omet and Abu-

Ruman (2003), and the Ben, Naceur and Goaied (2002) studies, their results should be treated

with caution.

Despite all the empirical work testing the DIH, the impact of dividend policy on the value of

a firm remains unresolved. In an earlier section, it was noted that the proposition of dividend

irrelevancy was based on several binding assumptions about the nature of perfect capital

markets. This is an „a priori‟ model of how markets should work if they were perfect.

Naturally, once we depart M&M‟s world of prefect capital market and relax one or more of

the assumptions of perfect capital markets, the issue of dividend policy becomes more

complicated. Introducing market imperfections might change the view that dividend decision

is irrelevant. Importantly, if dividend policy is relevant it may interact with other decisions

made by the firm about investment and financing. In other words, there may be a range of

reasons why dividend policy might matter as would be discussed in the following sections.

44

3.4 Bird-In-The-Hand Hypothesis (High Dividends Increase Stock Value):

Empirical Evidence

One alternative and older view about the effect of dividend policy on a firm‟s value is that

dividends increase firm value. In a world of uncertainty and imperfect information, dividends

are valued differently to retained earnings (or capital gains). Investors prefer the “bird in the

hand” (BIH) of cash dividends to the “two in the bush” of future capital gains. Increasing

dividend payments, ceteris paribus, may then be associated with increases in firm value.

As a higher current dividend reduces uncertainty about future cash flows, a high payout ratio

will reduce the cost of capital, and hence increase share value. That is, according to the so-

called “bird-in-the hand” hypothesis, high dividend payout ratios maximize a firm‟s value. A

pound of dividends has, on average, four times the impact on stock prices as a pound of

retained earnings and many studies supported the BIH hypothesis (Diamond, 1967. See also

Gordon and Shapiro, 1956; Gordon, 1959, 1963; Lintner, 1962; Walter, 1963).

M&M (1961) criticized the BIH hypothesis and argued that the firm‟s risk is determined by

the riskiness of its operating cash flows, not by the way it distributes its earnings.

Consequently, M&M called this argument the bird-in-the-hand fallacy. Further, Bhattacharya

(1979) suggested that the reasoning underlying the BIH hypothesis is fallacious. Moreover,

he suggested that the firm‟s risk affects the level of dividend not the other way around. That

is, the riskiness of a firm‟s cash flow influences its dividend payments, but increases in

dividends will not reduce the risk of the firm.

The notion that firms facing greater uncertainty of future cash flow (risk) tend to adopt lower

payout ratios seems to be theoretically plausible (see, for example, Friend and Puckett, 1964).

Empirically, Rozeff (1982) found a negative relationship between dividends and firm risk.

That is, as the risk of a firm‟s operations increases, the dividend payments decrease (see also

Jensen, Solberg, and Zorn, 1992).

Gordon (1959) hypothesized three reasons why investors would buy a certain stock. First, to

obtain both dividends and earnings, second, to obtain dividends, and finally to get the

45

earnings. He examined these hypotheses by estimating different regression models using

cross-section sample data of four industries (chemicals, foods, steels, and machine tools) for

two years 1951 and 1954. The dividend hypothesis was tested using a linear regression,

which is similar to the following equation,

Pit = α0 + α1Dit + α2Rit + εit………………………………………………………………………… 11

where, for firm i and period t, P,D, & R are the share price, dividends, and the retained

earnings, respectively. The reciprocal of the dividend coefficient α1 is the estimated required

rate of return on common stocks without growth, and the coefficient on retained earnings α2

is the price for growth. Gordon (1959) found that dividends have greater influence on share

price than retained earnings. In addition, he argued that the required rate of return on a share

increases with the fraction of retained earnings because of the uncertainty associated with

future earnings. Similarly, Gordon (1963) argued that higher dividend payouts decrease the

cost of equity or the required rate of return on equity. Using British data for the period

between 1949 and 1957, Fisher (1961) reached a similar conclusion that dividends have

greater impact on share prices than retained earnings.

The above regression (equation 11) was subject to several criticisms. Firstly, it does not take

into account the risk variation among firms drawn from different industries, and this may lead

to an upward bias in the coefficient on dividends α1. That is, high risk associated with a stock

may result in low price and low payout, while low risk associated with a stock may result in

high payout and low price. Secondly, the equation accounts only for the growth coming from

investments that are financed with retained earnings, while it ignores the growth that may

come from the use of external financing.

This may bias the coefficient of retained earnings α2. Thirdly, since dividends are more stable

than reported earnings, the short-run fluctuations in income will be mainly reflected in

change in retained earnings. If share prices and dividends are related to normal rather than

reported income, the equation is biased in favour of dividends. Finally, dividends are

measured more precisely than retained earnings because the estimated retained earnings

depend on the accounting procedure followed to measure total earnings, which will place an

46

additional downward bias on the retention coefficient α2 (see Friend and Puckett, 1964, and

Diamond, 1967).

To correct for the potential bias resulting from the criticisms mentioned above (especially 1

and 2), Diamond (1967) introduced into the regression equation (11) the average three-year

earning-price ratio centred on t -1. He examined the impact of dividends and retained

earnings on share prices for a sample of 255 US firms from eight industries for 1961 and

1967. He found only weak support for the notion that investors have preference for dividends

over retained earnings. The study found that in industries where rates of growth were

relatively high, retained earnings were marginally preferred more than dividends, whereas in

mature industries with low growth rate a dollar of dividends is slightly preferred to a dollar of

retained earnings. This suggests a negative relationship between a firm‟s growth and dividend

payout. The results obtained by Diamond (1967) are consistent with earlier findings of Friend

and Puckett (1964). Recently, Fama and French (2001) also found that firms with higher

growth and investments tended to have lower payouts.

Baker, Powell and Veit (2002a) surveyed managers of NASDAQ firms to assess their view

about dividend policy issues including the BIH hypothesis. Their questionnaire contains one

statement about the BIH hypothesis, stating “investors generally prefer cash dividends today

to uncertain future price appreciation”. Based on 186 responses, only 17.2 percent agree with

the statement, 28 percent no opinion, and 54.9 percent disagree. Therefore, they conclude,

“…this finding does not provide support for the bird-in-the-hand explanation for why

companies pay dividends” (p.278). In an earlier survey, Baker and Powell (1999) also found

a similar rate of agreement (17.7 percent) about the bird-in-the hand explanation of dividend

relevance.

Empirical support for the BIH hypothesis as an explanation for paying dividends is generally

limited, and the argument has been challenged especially by M&M (1961) who argued that

the required rate of return (or the cost of capital) is independent of dividend policy,

suggesting that investors are indifferent between dividends and capital gains. Indeed, based

on the tax-preference explanation, discussed shortly, researchers such as Litzenberger and

Ramaswamy (1979), among others, developed an explanation of dividend policy that reaches

47

the opposite result. That is, investors are disadvantaged in receiving cash dividends. The next

section examines the argument that low dividends are preferred to higher dividends.

3.5. Tax-Effect Hypothesis and Evidence on whether Low Dividend increases

Stock Values.

The M&M assumptions of a perfect capital market exclude any possible tax effect. It has

been assumed that there is no difference in tax treatment between dividends and capital gains.

However, in the real world taxes exist and may have significant influence on dividend policy

and the value of the firm. In general, there is often a differential in tax treatment between

dividends and capital gains, and, because most investors are interested in after-tax return, the

influence of taxes might affect their demand for dividends. Taxes may also affect the supply

of dividends, when managers respond to this tax preference in seeking to maximize

shareholder wealth (firm value) by increasing the retention ratio of earnings.

The tax-effect hypothesis suggests that low dividend payout ratio lowers the cost of capital

and increase the stock price. In other words low dividend payout ratios contribute to

maximising the firm‟s value. This argument is based on the assumption that dividends are

taxed at higher rates than capital gains. In addition, dividends are taxed immediately, while

taxes on capital gains are deferred until the stock is actually sold. This tax advantage of

capital gains over dividends tend to influence investors, to prefer companies that retain most

of their earnings rather than pay them out as dividends, and are willing to pay a premium for

low-payout companies. Therefore, a low dividend payout ratio will lower the cost of equity

and increases the stock price. Note that, this prediction is almost the exact opposite of the

BIH hypothesis, and of course challenges the strict form of the DIH.

In many EU countries and the US, a higher tax rate is applied to dividends as compared to

capital gains taxes7. Therefore, investors in high tax brackets might require higher pre-tax

7 Note that in the US, the Tax Reform Act of 1986 substantially eliminated the preferential tax treatment of

capital gains over dividends. Prior to the Act, dividends were taxed at a higher rate than capital gains, whereas

following the Act, dividends and capital gains were taxed at the same rate (see, Rozeff, 1982 and Casey &

Dickens, 2000).

48

risk-adjusted returns to hold stocks with higher dividend yield. This relationship between pre-

tax returns on stocks and dividend yields is the basis of a posited tax-effect hypothesis.

Brennan (1970) developed an after-tax version of the capital asset pricing model (CAPM) to

test the relationship between tax risk-adjusted returns and dividend yield. Brennan‟s model

maintains that a stock‟s pre-tax returns should be positively and linearly related to its

dividend yield and to its systematic risk. Higher pre-tax risk adjusted returns are associated

with higher dividend yield stocks to compensate investors for the tax disadvantages of these

returns. This suggests that, a stock with higher dividend yield will sell at lower prices because

of the disadvantage of higher taxes associated with dividend income. The Brennan model can

be described as:

E (Rit – Rft) = γ0 + γ1βit + γ2(Dit – Rft)…………………………………………….12

Where, Rit is the return on stock i in period t, Rft is the risk-less rate of interest, βit is beta

coefficient for stock i in period t (systematic risk), Dit and is the dividend yield of stock i in

period t. It is assumed that the coefficient γ2 is interpreted as an implicit tax bracket and is

independent of the level of the dividend yield D. If the coefficient of dividend yield (γ2) is

statistically different from zero and positive, the results are interpreted as evidence of a tax

effect. That is, higher pre-tax risk-adjusted returns are necessary to compensate investors for

holding high-dividend-paying stocks because of the disadvantage associated with dividend

income.

A large body of empirical research has been devoted to testing Brennan‟s model and to

understanding the relationship between dividend yields and stock returns. For example, Black

and Scholes (1974) tested Brennan‟s model and found no evidence of a tax effect and

concluded that low or high-dividend yield stocks do not affect the returns of stocks either

before or after taxes.

However, Litzenberger and Ramaswamy (1979) strongly challenged the results of Black and

Scholes and criticised their methods, especially their definition of dividend yield. Recall that

Black and Scholes used a long-term dividend yield definition (previous year‟s dividends

divided by previous year‟s closing share price). Litzenberger and Ramaswamy extended

49

Brennan‟s (1970) model and used a monthly dividend yield definition in classifying stock

into yield classes, a positive dividend-yield class and zero dividend-yield class.

The results of Litzenberger and Ramaswamy show that the coefficient on dividend yield

variable (γ2) is positive and highly significant. Blume (1980) reported positive and significant

dividend yield coefficient, using long-run measure of yield and therefore, provides empirical

support for Brennan (1970) model. Litzenberger and Ramaswamy (1979) thus conclude that

every dollar increase in return in the form of dividends, investors requires an additional 23

cents in before tax returns. The dividend coefficient (γ2) obtained by Litzenberger and

Ramaswamy is consistent in magnitude with that reported by Black and Scholes (1974). The

implication of Litzenberger and Ramaswamy‟s findings is that firms could increase their

share prices by reducing dividends. However, if this prediction holds, one may raise a

question about why corporations pay dividends at all.

Miller and Scholes (1982) challenged Litzenberger and Ramaswamy‟s conclusion, and

criticised their short-term (monthly) definition of dividend yield. They suggested that tests

employing a short-term dividend yield definition are inappropriate for detecting the impact of

differential tax treatment of dividends and capital gains on stock returns. Furthermore, Miller

and Scholes argued that the positive yield-return relation was caused by information bias. The

reason for this argument is that Litzenberger and Ramaswamy ignored the information effect

of dividend omissions. An announcement of dividend omissions (perceived as bad news) may

result in an upward bias in the dividend yield coefficient, since it reduces the return of the

zero yield-dividend class. Miller and Scholes attempted to correct the information bias and

then retest the Litzenberger and Ramaswamy result. They found that the dividend yield

coefficient was not statistically different from zero. Hess (1981) found similar results to

Miller and Scholes. In his study, Hess tested the relation between the monthly stock returns

and dividend yield over the period of 1926 to 1980. He found mixed results and concluded

that the work reinforces the findings of Miller-Scholes study and lends further empirical

support to the original M&M proposition.

Litzenberger and Ramaswamy (1982) re-examined the relationship between dividend yield

and stock returns after adjusting the dividend yield coefficient for any potential information

effects. Their results, consistent with their previous findings, were that the yield coefficient is

50

positive and statistically significant. Kalay and Michaely (2000) re-examine the Litzenberger

and Ramaswamy (1979) experiment using weekly data. They attempt to find whether the

positive dividend yield obtained by Litzenberger and Ramaswamy is due to tax effects or to

the information effects as conjectured by Miller and Scholes (1982). Kalay and Michaely

exclude all weeks containing dividend omissions. They find a positive and significant

dividend yield coefficient, inconsistent with Miller and Scholes‟s conjecture that the positive

yield coefficient is driven by information biases. Also, Morgan (1982) had shown evidence

inconsistent with the Miller and Scholes conjecture. Furthermore, using daily and monthly

British data, Poterba and Summers (1984) provide evidence that strongly supports the tax-

effect hypothesis.

Along the lines of Litzenberger and Ramaswamy (1979) and Blume (1980), Keim (1985)

used the Sharpe-Lintner CAPM to estimate the relation between long-run dividend yields and

stock returns among US firms. He used a sample of 429 firms in January 1931 and 1289

firms in December 1978. In his study, Keim constructed six dividend-yield portfolios. The

first portfolio contained all zero-dividend firms, and the other five ranked from lowest to

highest positive dividend-yield firms. He documented a non-linear relation between dividend

yields and stock returns, and his results rejected the hypothesis that average returns are equal

across portfolios.

In addition, Keim (1985) tested the impact of firm size and stock return seasonality on the

relationship between stock returns and dividend yields. He found a positive and significant

yield coefficient. However, much of the non-linear relation was concentrated in the month of

January for small firms. Nonetheless, the same result was obtained even after controlling for

firm size. In addition, he reported an inverse relationship between positive yield and firm size

as measured by market capitalization and concluded that the results suggest the observed

relation between long-run dividend yields and stock returns may not be solely attributable to

difference in marginal tax rates for dividends and capital gains. His results suggest a yield-

related tax effect. However, because of the significant effect of the month of January

(seasonality) on the relation between dividend yields and stock returns these findings are not

totally consistent with the after-tax CAPM. This conclusion deepens the puzzle surrounding

the issue of a yield-related tax effect.

51

In the UK, Morgan and Thomas (1998) examined the relationship between dividend yields

and stock returns over the period 1975 to 1993. Drawing on Keim‟s (1985) methodology,

they tested the tax-based hypothesis in which dividend yields and stock returns are positively

related. However, they pointed out that under the 1973 imputation tax system capital gains

received a disadvantaged tax treatment when compared to dividend income. Consequently the

tax-based hypothesis, in the UK, predicted a negative relation between dividend yields and

risk-adjusted stock returns. Stocks with low yields are expected to produce higher returns to

compensate stockholders for the increased tax burden associated with capital gains, and vice

versa.

Contrary to prediction, Morgan and Thomas find a positive relationship between dividend

yields and stock returns. Moreover, their results suggest a non-linear relation between risk-

adjusted returns and dividend yield, which is inconsistent with Brennan‟s model. Also, the

firm size and seasonality seems to influence the relationship between dividend yield and

stock returns. Morgan and Thomas were therefore unable to provide support for the tax-effect

hypothesis. In a previously mentioned study, Baker et al. (2002a) surveyed the managers of

630 NASDAQ firms and found weak or no support for the tax-preference theory.

The tax-effect hypothesis (TEH) is therefore based on a simple proposition. Many investors

are faced with dividends being taxed at a higher rate than capital gains. In addition, dividends

are taxed immediately, while taxes on capital gains are deferred until the gains are actually

realized. Therefore, the TEH suggests that taxable investors will demand superior pre-tax

returns from stocks that pay a large proportion of their income in the form of highly taxed

dividends. In other words, investors will value the dollar of capital gains greater than a dollar

of dividends, resulting in lower dividend-stocks selling at a relative premium to their higher-

dividend counterparts. From the argument above, the evidence with respect to the TEH

appears to be inconclusive.

In all of the studies discussed above the TEH has been addressed from one perspective: the

relationship between dividend yields and the stock returns (CAPM-based studies). The

literature, however, has also provided a vast amount of empirical research on the TEH by

examining the behaviour of stock prices around the ex-dividend day (ex-dividend day

studies). This issue will be further examined in the next section.

52

3.6 Clientele Effects of Dividends Hypothesis

In their seminal paper M&M (1961) noted that the pre-existing dividend clientele effect

hypothesis (DCH) might play a role in dividend policy under certain conditions. They pointed

out that the portfolio choices of individual investors might be influenced by certain market

imperfections such as transaction costs and differential tax rates to prefer different mixes of

capital gains and dividends. M&M argued that these imperfections might cause investors to

choose securities that reduce these costs. M&M termed the tendency of investors to be

attracted to a certain type of dividend-paying stocks a „dividend clientele effect‟. Similarly,

Pettit (1977) describes the net tendency of an individual investor to hold portfolios of

securities that have particular dividend paying characteristics as „dividend clientele effect‟.

Nonetheless, M&M maintained that even though the clientele effect might change a firm‟s

dividend policy to attract certain clienteles, in a perfect market each clientele is “as good as

another”. Hence the firm valuation is not affected as the dividend policy remains irrelevant.

In practice, investors often face different tax treatments for dividend income and capital

gains, and incur costs when they trade securities in the form of transaction costs.

For these reasons and based on different investors‟ situations, taxes and transaction costs may

create investor clienteles, such as tax minimisation induced clientele and transaction cost

minimisation induced clientele respectively. Another possible dividend clientele effect is

related to risk clienteles. High-payout stocks tend to be less risky than low-payout stocks;

and based on the risk factor, dividends may attract certain clientele investors ( Scholz, 1992).

These clienteles will be attracted to firms that follow dividend policies that best suit their

particular situations. Similarly, firms may tend to attract different clienteles by their dividend

policies. For example, firms operating in high growth industries that usually pay low (or no)

dividends attract a clientele that prefers price appreciation (in the form of capital gains) to

dividends. On the other hand, firms that pay a large amount of their earnings as dividends

attract a clientele that prefers high dividends.

53

Allen, Bernardo and Welch (2000) suggest that clienteles such as institutional investors tend

to be attracted to invest in dividend-paying stocks because they have relative tax advantages

over individual investors. These institutions are often subjected to restrictions in institutional

charters which, to some extent, prevent them from investing in non-paying or low-dividend

stocks. Similarly, good quality firms prefer to attract institutional clienteles (by paying

dividends) because institutions are better informed than retail investors and have more ability

to monitor or detect firm quality. This prediction is consistent with the signalling hypothesis.

In fact, Allen, Bernardo and Welch (2000) proposed a model that links signalling and agency

arguments with the clientele effects of dividend hypothesis. Allen et al, (2000) conclude with

the proposition that the clientele effects are the very reason for the presence of dividends.

3.6.1 The Tax-Induced Clientele-Effects

Since most of the investors are interested in after-tax returns, the different tax treatment of

dividends and capital gains might influence their preference for dividends or capital gains.

This is the essence of the tax-induced DCH. For example, ceteris paribus, investors in low

tax brackets who rely on regular and steady income will tend to be attracted to firms that pay

high and stable dividends. In addition, some corporate or institutional investors tend to be

attracted to high-dividend stocks (Han, Lee and Suk, 1999 and Dhaliwal, Erickson and

Trezevant, 1999).

Short, Zhang and Keasey, (2002) used UK data to provide strong evidence in support of the

notion that payout ratios and institutional ownership are positively related. However, they did

not link their results to the clientele hypothesis. On the other hand, investors in relatively high

tax brackets might find it advantageous to invest in companies that retain most of their

income to obtain potential capital gains, all else being equal. Some clienteles, however, are

indifferent between dividends and capital gains such as tax exempt and tax deferred entities.

54

3.6.2 The Transaction Cost-Induced Clientele and the Empirical Evidence

Another argument of the DCH is based on the proposition that dividend policy may influence

different clienteles to shift their portfolio allocation, resulting in transaction costs. For

example, small investors (such as retirees, income-oriented investors, and so on) who rely on

dividend income for their consumption needs, might be attracted to (and even may pay a

premium for) high and stable-dividend stocks, because the transaction costs associated with

selling stocks might be significant for such investors. On the other hand, some investors (e.g.

wealthy investors), who do not rely on their share portfolios to satisfy their liquidity needs,

prefer low payouts to avoid the transaction costs associated with reinvesting the proceeds of

dividends, which they actually do not need for their current consumption (Bishop et al.,

2000). Note that for both groups of investors, transforming one financial asset to another

incurs a transaction costs. Thus, the M&M‟s notion of homemade dividends is not costless

and the existence of such costs may make dividend policy not irrelevant.

The other effect of transaction costs on dividend policy is related to the fact that firms may

need to restore cash paid out as dividends with new equity issues (or debt financing) to take

advantage of new investment opportunities. If issuing costs are significant, then firms are

most likely to rely on retained earnings rather than external financing. This is reinforced by

the empirical fact that retained earnings constitute the major source of firm finance not just

in developing but also even in developed capital markets.

Fazzari, Hubbard and Petersen (1988) reported that, over the period of 1970 to 1984, the

retained earnings amounted to 71.1 percent of the total source of funds of US manufacturing

firms with an average retention ratio of 60 percent8. In these cases, there is a negative

relationship between transaction costs and dividend payments. Firms can reduce or avoid

such expenses by lowering dividend payments or not paying them at all. However, in

practice, many firms continue to pay cash dividends, while at the same time issuing new

equity and debt, suggesting that other factors may also be at work in influencing dividend

policy.

8 Note that the average retention ratio varies with the firm size. Smaller firms have greater retention ratio than

larger firms.

55

An important implication of the DCH is that, by changing its dividend policy, a firm‟s

ownership structure might also change. Another implication of clientele theory is that firms

should attempt to adopt a stable dividend policy to avoid inducing shareholders to modify

their portfolios, entailing transaction costs (Scholz, 1992 and Soter, Brigham and Evanson,

1996). However, whether these changes have an impact on firm value remains unanswered.

In a competitive equilibrium, it would be expected that one clientele is good as another and

that the clientele effect of dividend policy is irrelevant to a firm‟s value. To date, evidence on

the link between clientele effects and firm value is inconclusive.

The theoretical plausibility of dividend clientele hypothesis is relatively ambiguous. On the

one hand, transaction costs and taxes may influence demands for dividends. But the mere

existence of transaction costs or differential taxes is not enough rationale for a general

theoretical explanation of the determination of dividend policy. Not surprisingly, therefore,

most of the literature on the DCH has produced mixed results. Miller (1977) and Auerbach

(1983) have presented theoretical models that are consistent with the presence of clientele

effects.

A number of empirical studies discussed have examined investors‟ portfolios, and their

demographic attributes including taxes. Pettit (1977) provided empirical evidence for the

existence of a clientele effect by examining the portfolio positions of 914 individual

investors. He found a significant positive relationship between investors‟ ages and their

portfolios‟ dividend yield, and a negative relationship between investors‟ incomes and

dividend yield. Pettit suggested that elderly low-income investors tend to rely more on their

portfolios to finance their current consumption, and avoid the transaction costs associated

with selling stocks. Consequently, they have more of a tendency to invest in high-dividend

stocks. Pettit also showed that investors whose portfolios have low systematic risk prefer

high-payout stocks, and thus found evidence for tax-induced clientele effect.

However, using a sample constructed from the same database used in Pettit‟s (1977) study,

Lewellen et al. (1978) found only very weak supportive evidence of the clientele effect

hypothesis. In a later study, Scholz (1992) developed an empirical model to test the DCH

directly by examining individual investor portfolio data. He found that differential tax

56

treatment of dividends and capital gains influences investors‟ decisions in choosing between

higher-or-lower-dividend yield portfolios, consistent with dividend tax–clientele hypothesis.

Another strand of empirical testing has examined the relationship between dividend changes

and clientele effect. Richardson, Sefcik and Thompson (1986) tested a sample of 192 US

firms that initiated dividends for the first time during the period of 1969 through 1982. They

attempted to investigate whether the observed (post-dividend-initiations) increase in firms‟

stocks trading volume is due to the signalling effect or a product of investors in various tax

clienteles adjusting their portfolios. They found that the increased trading volume associated

with dividend policy changes was mainly related to the information contained in the dividend

announcement, and only a small part was related to clientele adjustment. They concluded that

the evidence supporting the existence of clientele trading is somewhat weak. This concurs

with Asquith and Krasker (1985) who had similar findings.

Dhaliwal, Erickson and Trezevant (1999) examined institutional shareholding changes

following dividend initiations. Based on the theory of tax-induced clienteles, they expected

an increase in institutional ownership subsequent to dividend initiations. Using a sample of

133 dividend initiators from the 1982 to 1995 period, the results obtained are consistent with

their prediction. They reported an 80% significant increase in their sample firms who

experienced an increase in institutional shareholders following dividend initiation. They

concluded that the dividend/tax-clientele effect is strong enough to influence investors‟

decisions.

Seida (2002) also provided evidence consistent with the DCH findings. Earlier research by

Bajaj and Vijh (1990), Ang, Blackwell and Megginson (1991), and Denis, Denis and Sarin

(1994) provided empirical support for the existence of the dividend clientele hypothesis.

Michaely, Thaler and Womack (1995), however, found weak or no empirical support for this

notion. For example, they tested the possible changes in shares turnover rate for firms

announcing dividend initiation or omission as a sign of a clientele change, and found little

evidence.

57

Finally, another strand attempts to infer the tax characteristics of a firm‟s marginal investors

by examining the movements of stock prices around the ex-dividend days9, and therefore

provides an indirect test of the DCH. The basic intuition of the relation between stock price

and ex-dividend day is that, in a rational capital market, and in a world of certainty, share

prices should drop by approximately the amount of dividend per share on the day the stock

goes ex-dividend. Any other share price behaviour suggests arbitrage opportunities ( Kalay,

1982 and Lease et al., 2000). When the stock goes ex (without)-dividend the investor has no

claim to dividend payments, and thus will not pay the same amount as if the stock traded cum

(with)-dividend. The stock price on the ex-dividend day should therefore be lower than in the

cum-dividend period to reflect the lost dividend (Lease et al., 2000).

This notion, however, may not perfectly hold in some circumstances, because dividends are

usually taxed more heavily than capital gains. Investors in high tax brackets will therefore be

better off receiving their income in the form of capital gains rather than dividends. The tax

effect may mean that the drop in stock price may be less than the dividend because investors

value dividends less than capital gains. Elton and Gruber (1970) had presented empirical

evidence about the tax-induced clientele hypothesis by observing the share price behaviour

around the ex-dividend day10

.

They examined the shares listed on the NYSE paying a dividend between April 1, 1966 and

March 31, 1967, Elton and Gruber found that share prices fell by less than the amount of the

dividend on ex-dividend days. This is because the positive relationship between the dividends

yields of a stock and the proportionate size of its ex-dividend price drop. They interpreted

their results as evidence that differential taxes induced a preference for capital gains relative

to cash dividends, therefore supporting the tax clientele hypothesis (that is, investors in high

tax brackets invest in low-dividend yield stocks and vice versa). They thus concluded that

9 The ex-dividend day is the first day in which the stock is traded “without” dividends, i.e. the current dividend

is earmarked for the seller, not the buyer. The cum-dividend day is the last business day the stock is traded

“with” dividend i.e. the stock‟s buyer is entitled to receive a declared dividend

10 Several empirical studies have analysed ex-dividend day behaviour of share prices for US firms. A partial

listing includes Kalay (1982, 1984), Elton, Gruber and Rentzler (1984), Eades, Hess and Kim (1984), Barclay

(1987), Karpoff and Walkling (1988, 1990), Grammatikos (1989), Bali and Hite (1998), Koski and Scruggs

(1998), and Elton, Gruber and Blake (2003).

58

firms do not only seem to attract a clientele but they attract a rational clientele (one which

should prefer their dividend policy).

Kalay (1982) criticised Elton and Gruber and argued that the marginal tax rates of the

investors cannot be inferred from the ex-dividend day price-drop-to-dividend ratio (price-

drop ratio)11

, and the observed positive relationship between price-drop ratio and dividend

yield may not be due to tax induced clientele effects. He presented another explanation,

known as the “short-term traders” hypothesis. Kalay argued that, assuming certainty, if the

ex-dividend price ratio drop is less than one (less than the amount of dividends), short-term

traders who face the same tax rate on dividends and capital gains could make arbitrage

profits. That is, investors can buy a stock before it goes ex-dividend and sell it soon after.

This practice is known as dividend capture (Karpoff and Walkling, 1990).

However, this arbitrage process could be hampered by transaction costs12

. Kalay suggests that

transaction costs are insignificant for broker dealers who are the potential short-term traders,

while Elton, Gruber and Rentzler (1984) argue that it matters even for broker dealers. Karpoff

and Walkling (1988, 1990) show that excess ex-dividend-day returns are positively correlated

with transaction costs (measured by bid-ask spread), and this relationship increases for stocks

with high-dividend yields. They suggest that short-term trading around ex-dividend days is

higher for high-yield stocks (see also Michaely and Vila, 1996), implying that short term

trading (or dividend capture) may influence the ex-dividend day stock price changes, and

hence any clientele effects may not be the only explanation for these changes.

Examination of the ex-dividend day behaviour of share prices has also been extended to

different stock markets, including Australia (Brown and Walter 1986), Canada (Lakonishok

and Vermaelen, 1983 and Booth and Johnston, 1984), Finland (Hietala, 1990), Greece

11

The price-drop ratio can be defined as (PB-PA)/D, where, PB is the stock price cum-dividend, PA is the stock

price on the ex-dividend day, and D is the amount of dividends.

12 The arbitrage profit can also be inhibited by the risk or the uncertainty of ex-dividend price. For further

details on this issue see for example, Heath and Jarrow (1988), Grammatikos (1989). More recently, Michaely

and Vila (1996) documented that risk and transaction costs affect the abnormal trading volume adversely.

59

(Milonas and Travlos 2001), Japan (Kato and Loewenstein, 1995), and New Zealand

(Bartholdy and Brown, 1999). These studies have found mixed support for the ex-dividend

day effect.

It is worth noting that dividend clientele hypothesis predictions, to some extent, may

contradict other explanations of dividend policy such as the signalling and agency costs

hypotheses, discussed above. For example, according to the signalling hypothesis, dividends

convey information about a firm‟s future prospects, and in that sense investors with

preferences for capital gains (for tax reasons) may still prefer firms with high-payout ratios,

contradicting the prediction of the tax-induced clientele hypothesis. Also, based on agency

theory, dividends may mitigate the free cash in hand of managers and reduce the agency

problems, and for these reasons investors may also prefer high-dividend stocks even though

they are tax-disadvantaged.

3.7 The Signalling Hypothesis and the Empirical Evidence

The signalling hypothesis (hereafter referred to as information content of dividend

hypothesis) is another reason why M&M‟s DIH is inadequate as an explanation of financial

market practice. The ICD hypothesis explains the asymmetric information between insiders

(managers and directors) and outsiders (shareholders). M&M assumed that managers and

outside investors have free, equal and instantaneous access to the same information regarding

a firm‟s prospects and performance. But managers who manage the firms usually possess

information about its current and future prospects that is not available to outsiders. This

informational gap between insiders and outsiders may cause the true intrinsic value of the

firm to be unavailable to the market. If so, share price may not always be an accurate measure

of the firm‟s value.

In an attempt to close this gap, managers may need to share their knowledge with outsiders so

they can more accurately understand the real value of the firm. Historically, due to a lack of

complete and accurate information available to shareholders, the cash flow provided by a

security to an investor often formed the basis for its market valuation (Baskin and Miranti,

1997). Thus, dividends provide a useful tool for managers in which to convey their private

60

information to the market because investors used visible (or actual) cash flows to equity as a

way of valuing a firm. Many academics and financial practitioners also suggest that dividends

might have implicit information about a firm‟s prospects.

Interestingly, M&M (1961) suggested that when markets are imperfect share prices may

respond to changes in dividends. This implies that dividend announcements may be seen to

convey implicit information about the firm‟s future earnings potential. The proposition has

since become known as the “information content of dividends” or signalling hypothesis.

However, M&M dismissed the possibility that this occurred by suggesting that the empirical

evidence does not support the notion that investors prefer dividends to retained earnings.

According to the signalling hypothesis, investors can infer information about a firm‟s future

earnings through the signal coming from dividend announcements, both in terms of the

stability of, and changes in, dividends. However, for this hypothesis to hold, managers should

first possess private information about a firm‟s prospects, and have incentives to convey this

information to the market. Second, a signal should be true; that is, a firm with poor future

prospects should not be able to mimic and send false signals to the market by increasing

dividend payments. Thus the market must be able to rely on the signal to differentiate among

firms. If these conditions are fulfilled, the market would react favourably to the

announcements of dividend increase and unfavourably if otherwise (Ang, 1987, and Koch

and Shenoy, 1999).

As managers are likely to have more information about the firm‟s future prospects than

outside investors, they may be able to use changes in dividends as a vehicle to communicate

information to the financial market about a firm‟s future earnings and growth. Outside

investors may perceive dividend announcements as a reflection of the managers‟ assessment

of a firm‟s performance and prospects. An increase in dividend payout may be interpreted as

the firm having good future profitability (good news), and therefore its share price will react

positively. Similarly, dividend cuts may be considered as a signal that the firm has poor

future prospects (bad news), and the share price may then react unfavourably.

Accordingly, it would not be surprising to find that managers are reluctant to announce a

reduction in dividends. Lintner (1956) argued that firms tend to increase dividends when

61

managers believe that earnings have permanently increased. This suggests that increase in

dividend implies long-run sustainable earnings. This prediction is also consistent with what is

known as the “dividend-smoothing hypothesis”. That is, managers will endeavour to smooth

dividends over time and not make substantial increases in dividends unless they can maintain

the increased dividends in the foreseeable future. Lipson, Maquieira and Megginson (1998)

observed that, managers do not initiate dividend increase until they believe those dividends

can be sustained by future earnings. Fama and Babiak (1968), Baker, Farrelly, and Edelman

(1986), DeAngelo, DeAngelo, and Skinner (1992), and Baker, Veit and Powell (2001) find

support for Lintner‟s argument of dividend smoothing.

It is worth noting that, although management can use changes in dividends as a signal to

convey information to the market, in some cases dividend changes may be an ambiguous

signal. The most cited dividend signalling models can be found in Bhattacharya (1979), John

and Williams (1985), and Miller and Rock (1985). The literature of dividend signalling has

produced other models to represent the theory of asymmetric information (see, for example,

Ambarish, John and Williams, 1987, Ofer and Thakor, 1987, Kumar, 1988, and Bernheim,

1991, Allen, et al., 2000).

In general, these models are based on several assumptions. There is asymmetric information

between corporate insiders (managers) and outside investors (shareholders). Dividends

contain information about the firm‟s current and future cash flows, and managers have

incentives to convey their private information to the market through dividend payments in

order to close the information gap. The announcement of a dividend increase will be taken as

good news and the market will bid up share prices accordingly. Similarly, an announcement

that a dividend will be cut suggests unfavourable prospects and will tend to see the firm‟s

share price fall (Asquith and Mullins (1983) Kalay and Loewenstein (1985), Denis et

al.,1994, Yoon and Starks, 1995, and Bali,2003).

Dividends are considered a credible signalling device because of the dissipative costs

involved. For example, in Bhattacharya (1979) model the cost of signalling is the transaction

cost associated with external financing. In Miller and Rock (1985) model the cost is the

distortion in the optimal investment decision, whereas in the John and William (1985) model,

the signalling cost is the tax penalty on dividends relative to capital gains.

62

Therefore, only good-quality firms (under valued) can use dividends to signal their prospects,

and poor-quality firms cannot mimic by sending a false signal to the market because of the

costs involved in that action. A major criticism addressed to these models is that firms choose

dividends to signal their prospects while other less costly means are available such as share

repurchases (Allen and Michaely, 2002).

In the preceding section, the theory of dividend signalling was developed around the

proposition that corporate insiders are more informed about the firm‟s current performance

and future prospects than outsiders. This suggests that the market perceives dividends (and

repurchases) as signals of a management‟s view about the firm‟s fortunes, and therefore share

prices react to that signal. The empirical work on dividend signalling has examined two main

issues. First, whether share prices move in the same direction with dividend change

announcements. Second, whether dividend changes enable the market to predict future

earnings.

Finance scholars have addressed these issues extensively, but once again the results have

been mixed and inconclusive. The first question has received much attention in the literature,

because if the announcement of dividend changes does not have the predicted impact on

share prices this will cast doubt on the validity of the information content of dividend

hypothesis. Pettit (1972) observed that dividend announcements communicate valuable

information, and showed that the market reacts positively to the announcement of dividend

increases (significant increase in stock prices), and negatively to the announcement of

dividend decreases (significant drop in stock prices).

He argued that dividend announcement, when forthcoming, may convey significantly more

information than the information implicit in an earnings announcement. Aharony and Swary

(1980) suggest that dividend and earning announcements are not perfect substitutes and a

proper test for the signalling hypothesis needs to take into account the effect of earnings

announcements. They found support for the results obtained by Pettit even after controlling

for contemporaneous earnings announcements. Woolridge (1983) also found significant

increase (decrease) in common stock returns following the unexpected dividend increase

(decrease) announcements.

63

Asquith and Mullins (1983) examined the market‟s reaction to dividend announcements for a

sample of 168 firms that initiated dividends either for the first time in their corporate history

or resumed paying dividends after at least a ten-year period. They tested the average daily

excess stock returns ten days before and ten days after the announcement of dividend

initiation. For the two-day announcement period, their result shows that there is an excess

return of about +3.7 percent. Moreover, using cross-sectional regression Asquith and Mullins

found a positive and significant relationship between the magnitude of initial dividends and

the abnormal returns on the announcement day. This suggests that the size of dividend

changes may also matter. In another empirical study, Asquith and Mullins (1986) reinforce

their earlier findings and offer more support to the information content of dividend

hypothesis.

Michaely, Thaler and Womack (1995) have gone further by examining the impact of both

initiations and omissions of cash dividends on share prices reaction. They observed 561

dividend initiation events and 887 dividend omission events over the period of 1964 to 1988.

They documented that, during three days surrounding the announcements, the average excess

return was about –7.0 percent for omissions and +3.4 percent for dividend initiations. Note

that the market reactions to dividend omissions are greater than for dividend initiations.

This implies that the market reacts optimistically toward dividend initiations (or increases);

however, the market is more pessimistic in response to the announcements of dividend

omissions (or decreases). They found significant long-run drifts in stock prices in response to

dividend initiations and omissions and reported +7.5 percent excess returns after one year of

initiation announcements and +24.8 percent after three years. For dividend omissions they

reported abnormal returns of –11.0 percent in the first year and –15.3 percent after three

years. Bali (2003) presented evidence consistent with the preceding results. He reported an

average 1.17 percent abnormal return for dividend increases and -5.87 percent for decreases.

In addition, he examined the long run drifts of stock prices reaction to dividend increases and

decreases and reinforced the findings of Michaely et al. (1995).

From the empirical findings of these studies there seems to be general agreement that share

prices follow the same direction as the dividend change announcements. Dividend increases

64

and dividend initiations (decreases and omissions) are associated with subsequent significant

increases (decreases) in share prices. Moreover, the reaction of share prices in the event of

dividend decreases and dividend omissions is found to be more severe.

The signalling power of dividends, however, may vary in different markets. For example, in a

comparison study of dividend policies between Japanese and US firms, Dewenter and

Warther (1998) revealed that the influence of dividends as a signalling mechanism in Japan is

significantly lower as compared to the US. They studied 420 US firms and 194 Japanese

firms. The results of Dewenter and Warther‟s study show that the 2-day event window

(0, +1)13

in the event of dividend omissions gives a mean returns of -2.53 percent and -4.89

percent, while for dividend initiations +0.03 percent and +2.38 percent for Japanese and US

firms respectively. For a wide 62-day window (-60, +1) average returns are –6.48 percent and

-17.03 percent, while for dividend initiations +0.1 percent and +10.24 percent for Japanese

and US firms respectively.

The results indicate that the impact of dividend omission and initiation announcements on US

stock prices is significantly larger than on Japanese stock prices. Moreover, Dewenter and

Warther conclude that Japanese firms are subject to less information asymmetry especially

among keiretsu (industrial groups) member firms. These differences in the findings are

attributable to the differences in corporate governance structures between Japan and the US,

and moreover to the nature of corporate ownership in Japan. Conroy, Eades and Harris (2000)

provide evidence consistent with Dewenter and Warther‟s (1998) study for Japanese firms.

Using a sample of 200 German firms listed on Frankfurt Stock Exchange, Amihud and

Murgia (1997) found support for the notion that dividend changes convey information about

firms‟ values. They examined the stock price reaction to dividend announcements using 255

events of dividend increase and 51 events of dividend decrease for the period of 1988 to

1992, and compared the results with findings of studies based on US data. They reported that

the average abnormal return (AR) of stock prices is 96.5% percent for dividend increase and

–1.73 percent for dividend decrease.

13

Chapter two of the literature review on mergers and acquisitions and the research methodology (chapter 5),

have detailed discussions on event window.

65

In addition, they observed that though the earnings news preceded dividend change

announcements, dividends still have significant information. However, the finding of their

study was inconsistent with tax-based signalling models (for example, John and William,

1985, and Bernheim, 1991) because dividends in Germany are not tax-disadvantaged. Recall

that the tax-based signalling models propose that higher taxation on dividends makes them

informative about a firm‟s value. Thus, according to these models, if dividends do not suffer

from a tax penalty (as in the case Germany) share prices should not react to dividend

changes.

Travlos, Trigeorgis and Vafeas (2001) provided evidence from an emerging market in favour

of the dividend signalling hypothesis. They used a sample of 41 announcements of cash

dividend increase and 39 announcements of stock dividends for firms listed on the Cyprus

Stock Exchange for the period of 1985 to 1995, and examined market reaction to the

announcement of cash dividend increases and stock dividends. They found positive and

significant abnormal returns for both cash dividend increases and stock dividend

announcements and interpreted their results as consistent with the signalling hypothesis.

Numerous studies have addressed another question of the information content of dividends

hypothesis; that is, whether dividend changes enable the market to predict the future earnings

of a firm. Empirical work that addresses this issue has yielded puzzling results. For example,

Watts (1973) used a sample of 310 firms for the years 1946 to 1967, and annual definitions of

dividends and earnings to test the hypothesis that current and past dividends provide more

information to predict future earnings than that contained in current and past earnings. He

tested the relationship between annual future earnings in year and the level of dividends and

also examined the association between the abnormal increase/decrease in stock prices and

unanticipated changes in dividends. He found that the average estimated coefficients of

current dividends (across firms) are found to be positive but the average significance level are

too small and concluded that the information content of dividends are only be trivial.

Using a sample of 1025 firms listed on the NYSE and on the American Stock Exchange

(AMEX) between 1979 and 1991, Benartzi, Michaely and Thaler (1997) studied the

relationship between firms‟ future earnings and dividend changes. They did not find evidence

66

to support the notion that changes in dividends have the power to predict changes in future

earnings. They found that dividend changes are strongly linked to contemporaneous and

lagged earnings changes. And pose a challenge to the signalling hypothesis. Their results lend

support to the earlier findings of Watts (1973) and those of DeAngelo, DeAngelo, and

Skinner (1996) who also found no evidence that dividends provide valuable information

about future earnings.

Further, Nissim and Ziv (2001) found that dividend changes and earnings changes are

positively correlated, and provide support for the signalling hypothesis. However, their

results were not the same for dividend increases and decreases. They did not find an

association between dividend decreases and future profitability after controlling for current

and expected profitability, and they assumed that this result is possibly due to the accounting

conservatism. Note however that the proposition that dividend changes transmit information

about future earnings seems to have weak support.

Mixed support exists about issues relating to the information content of dividends hypothesis

which is a common concept in the dividend literature. Firms often use dividend policy to

communicate information about their future prospects to the market, and this provides

another possible explanation of why firms pay dividends. Moreover, signalling could play a

pivotal role in determining firms‟ dividend policies and their values.

.

The signalling hypothesis makes an important assumption that managers want to signal the

proper value of the firm via dividends. However, another school of thought has been

developed on the suspicion that managers may have incentives not to pay dividends and will

therefore need to be forced (or given incentives) to pay dividends.

3.8 Agency Costs and Free Cash Flow Hypothesis of Dividend Policy and the

Empirical Proof

One of the assumptions of M&M‟s perfect capital market is that there are no conflicts of

interests between managers and shareholders. In practice, however, this assumption is

questionable where the owners of the firm are distinct from its management. In these cases

managers are always imperfect agents of shareholders (principals). This is because managers‟

67

interests are not necessarily the same as shareholders‟ interests, and they might carry out

actions that are costly to shareholders, such as over-investing in managerially rewarding but

unprofitable activities. Shareholders therefore incur (agency) costs associated with

monitoring managers‟ behaviour, and these agency costs are an implicit cost resulting from

the potential conflict of interest among shareholders and corporate managers. The payment of

dividends might serve to align the interests and mitigate the agency problems between

managers and shareholders, by reducing the discretionary funds available to managers

(Rozeff, 1982, Easterbrook, 1984, Jensen, 1986, and Alli, Khan and Ramirez, 1993).

Another source of the agency costs problem that may be influenced by dividend policy is the

potential conflict between shareholders and bondholders. Shareholders are considered as the

agents of bondholders‟ funds. In this case, excess dividend payments to shareholders may be

taken as shareholders expropriating wealth from bondholders (Jensen and Meckling, 1976).

Shareholders have limited liability and they can access the company‟s cash flow before

bondholders. Consequently, bondholders prefer to put constraints on dividend payments to

secure their claims. Conversely, for the same reasons, shareholders prefer to have large

dividend payments (Ang, 1987).

Easterbrook (1984) argued that dividends could be used to reduce the free cash flow in the

hands of managers and hypothesised that dividend payments will oblige managers to

approach the capital market to raise funds. In this case investment professionals such as

bankers and financial analysts will also be able to monitor managers‟ behaviour. Therefore,

shareholders are able to monitor managers at lower cost (and minimise any collective action

problems). This suggests that dividend payments increase management scrutiny by outsiders

and reduce the chances for managers to act in their own self-interest. However, Easterbrook

suggested that increasing dividend payments might force managers to take undesirable

actions like increasing firm leverage, which may sometimes increase the riskiness of the firm.

Along the lines of Easterbrook‟s argument, Jensen (1986) contended that firms with excess

cash flow give managers more flexibility for using the funds in a way that benefit themselves

but not shareholders‟ best interests. He defined free cash flow as the cash not needed to

finance positive NPV projects and argued that managers have incentives to enlarge the size of

their firms beyond the optimal size. This is to amplify the resources under their control and

68

moreover to increase their compensation, which is often related to firm size (see also Gaver

and Gaver, 1993).

Thus, if a firm has a substantial surplus of cash the overinvestment problem will be more

pronounced, and managers may undertake negative NPV projects. Extracting the excess

funds of free cash flow that management controls can reduce this overinvestment problem.

Increasing dividend payouts may help to mitigate the free cash flow under managers‟ control,

thereby preventing them from investing in negative NPV or poor projects. As a result, paying

more dividends will reduce the agency costs between managers and shareholders. Moreover,

Jensen has pointed out that debt might play a similar role to dividends in reducing the agency

costs of free cash flow by reducing the funds under management control.

As noted earlier, M&M suggested that a firm‟s dividend policy is independent of its

investment policy. By contrast, the free cash flow hypothesis implies that dividend policy and

the investment decision are interrelated. It is argued that an increase in dividend payments

will reduce the “overinvestment” problem, which will have a positive impact on the market

value of the firm (Lang and Litzenberger, 1989). However, accepting the notion that

increasing dividends will reduce the funds available to managers and force them to be in the

market to acquire funds means that shareholders should be willing to tolerate the risk of the

firm being more indebted and also accept paying higher personal tax rates on dividends. In

other words, shareholders have to trade off between the costs and benefits of acquiring more

dividends.

The issue of agency costs hypothesis as an explanation of corporate dividend policy has been

widely addressed in empirical research. Rozeff (1982), for instance, was one of the first to

formally model agency costs using a large sample of US firms. Rozeff‟s regression model

and the hypothesised signs of the variables can be described as follows:

PAY = β0 + β1INS – β2GROW1 – β3GROW2 – β4BETA +β5STOCK + ε………………….13

where, PAY is the average payout ratio over a seven year period, INS is the percentage of

common stock held by insiders over the seven year period, GROW1 is the realized average

growth rate of a firm‟s revenues over a five year period, GROW2 is the forecasted growth of

69

sales over the five year period, BETA is the firm‟s estimated beta coefficient reported in the

Value Line Investment Survey, and STOCK is the natural log of the number of shareholders

at the end of the seven year period.

The key idea of Rozeff‟s (1982) model is that the optimal dividend payout is at the level

where the sum of transaction and agency costs are minimised, therefore the model is called

“cost minimisation model”. Rozeff‟s model contained two proxies for agency costs, namely

INS and STOCK. Note that the hypothesised signs of these variables (INS and STOCK) are

negative and positive, respectively. This indicates that there should be a negative relationship

between the percentage of stock held by insiders (insider ownership) and the payout ratio,

and a positive relationship between the number of shareholders (dispersion of ownership) and

the dividend payout ratio. Rozeff suggested that the benefits of dividends in reducing agency

costs are smaller for companies with lower dispersion of ownership and higher for insider

ownership. He found the agency costs variables significant and consistent with their

hypothesised sign. Rozeff‟s (1982) results provide empirical support for the agency costs

hypothesis. A decade later, Dempsey and Laber (1992) updated the work of Rozeff using an

extended period over the years 1981-1987 and strongly supported Rozeff‟s findings (see also

Lloyd, Jahera and Page, 1985).

Using factorial analysis to model the determinants of corporate dividend policy Alli et al.

(1993) found the ownership dispersion factor insignificant in relation to dividend decision,

inconsistent with Rozeff (1982). However, the insider ownership variable was found to be

significant and negatively related to dividend payouts. The overall results of their study

support the agency cost hypothesis of dividend policy. Jensen, Solberg and Zorn (1992)

applied three-stage least squares to examine the determinants of cross-sectional differences in

insider ownership, debt, and dividend policy.

They used a sample of 565 firms for the year 1982 and 632 firms for the year 1987. From the

dividend equation, the insider ownership variable was found statistically significant with a

negative sign. This implies that there is a negative relationship between insider holdings and

dividend payments. The result of Jensen et al. is consistent with Rozeff (1982) and therefore

supports the agency costs hypothesis.

70

Holder, Langrehr and Hexter (1998) examined 477 US firms over the period 1980 to 1990.

They reported that insider ownership and dividend payouts are significantly and negatively

related and that the number of shareholders positively influences payouts. In addition, they

found support for Jensen‟s free cash flow hypothesis. Likewise, Saxena (1999) examined a

sample of 235 unregulated and 98 regulated firms listed on the NYSE over the period of 1981

to 1990 and reinforced the findings of the Holder et al.‟s study. Both studies are consistent

with the agency costs hypothesis and provide evidence that agency cost is a key determinant

of the firm dividend policy.

Further empirical support for the agency cost hypothesis and in particular for the free cash

flow hypothesis came from Lang and Litzenberger (1989). While this study pre-dates many

of the agency costs studies discussed above, it does so from a different model of agency cost

analysis, namely the free cash flow hypothesis of Jensen (1986). Lang and Litzenberger

tested a sample of 429 dividend-change announcements for US firms for the period of 1979

to 1984 and used the Tobin‟s ratio14

to distinguish between overinvesting-firms and value-

maximising ones.

Based on Lang and Litzenberger (1989) overinvestment hypothesis, firms with less than one

should be expected to have larger average stock returns following dividend change

announcements. To clarify, low firms experience positive abnormal stock returns following

dividend increase announcements because the market anticipates this as a reduction in the

overinvestment problem (good news). That is, an increase in dividend payment reduces the

cash flow that would have been otherwise invested in negative NPV projects. Conversely,

dividend decrease suggests that the potential for overinvestment problems may have grown.

This prediction is consistent with the free cash flow hypothesis. They reported that the

average stock returns for firms were significant for both dividend increases and decreases.

They opined that dividend changes for over investing firms (where the Tobin‟s Q is less than

1) signal information about investment policies. This argument provides evidence in support

14 Tobin‟s Q can be defined as the ratio of the market value of the firm‟s equity and debt to the replacement cost

of its assets (Tobin, 1969 and Perfect & Wiles, 1994).

71

of free cash flow hypothesis and argued that the excess funds hypothesis provides a better

explanation of share price reaction to dividend change announcements than the cash flow

signalling hypothesis.

Other empirical studies have examined the agency theory of free cash flow and have found

little or no support for the excess cash flow hypothesis. For example, using a sample of 55

self-tenders and 60 special dividend announcements between 1979 and 1989 Howe, He and

Kao (1992) produced findings that challenge those of Lang and Litzenberger (1989) and

show that there is no relationship between Tobin‟s Q and stocks reaction to one-time

dividend announcements.

Further, Denis, Denis and Sarin (1994) investigated a sample of 5992 dividend increases and

785 dividend decreases between 1962 and 1988. They examined the relationship between

dividend yield and Tobin‟s Q. They found the relationship to be negative and argued that this

negative relation is attributable to a negative correlation between dividend yield and Tobin‟s

Q. This suggests that the market perceived this as a signal that overinvestment problems may

be imminent. In the same vein, they examined the level of capital expenditures for low and

high firms in relation to dividend changes and observed that firms with Tobin‟s Q less than 1

increased their investments following dividend increases and decreased them following

dividend decreases.

This result contradicts the overinvestment hypothesis. In addition, using a sample 419

dividend change announcements of firms listed on the NYSE over the period 1969 to 1988,

Yoon and Starks (1995) arrived at the same conclusion. The studies of both Denis et al.

(1994) and Yoon and Starks (1995) thus provide support to the cash flow signalling

hypothesis rather than free cash flow hypothesis as an explanation for the stock price

reactions to dividend change announcements.

Lie (2000) also examined the free cash flow hypothesis using a large sample of special

dividends, regular dividend increases, and self-tender offers. He found little evidence in

support of the agency cost hypothesis and concluded that neither small special dividends nor

the increase in regular dividends could solve the overinvestment problem. This is inconsistent

with the agency hypothesis of free cash flow. La Porta et al. (2000a) examined more than

72

4000 companies from 33 countries around the world and provided empirical support for the

agency costs hypothesis. They started by dividing the countries into two categories: countries

that provided good legal protection for minority shareholders, and countries where

shareholders had poor legal protection15

. Next, they analysed the effect of investor protection

on dividend payouts and tested two alternative agency models: the “outcome” model and the

“substitute” model. The first model implies that in countries with a more effective legal

protection system, shareholders have greater rights and therefore can force managers to

disgorge cash. As a result, dividend becomes an outcome of the legal protection of

shareholders. They hypothesised that the more effective the legal protection the greater the

rights of shareholders, and subsequently more dividends are paid.

The second or substitute model, predicts that managers can use dividends to establish a

reputation if they need to go to the capital market to raise external funds. In countries with

weak protection of shareholders, firms may need to establish a good reputation for their

treatment of investors by paying more dividends to shareholders. That is, dividends serve as a

substitute for legal protection of minority shareholders. La Porta et al. hypothesised that

higher payouts are expected in countries with poor legal protection and support the agency

model of dividends. That is, in countries where shareholders have better protection, firms pay

more dividends. Moreover, they found that firms operating in these countries and having a

rapid growth rate paid fewer dividends than their counterparts with slow growth rates. This

implies that shareholders use their legal power to force managers to disgorge cash when

investment opportunities are low.

Thus in summary, the empirical results for the agency costs explanation of dividend policy

are mixed. The agency costs hypothesis posits that dividends mitigate the cash under

management control, and therefore reducing the possibility that managers will use the funds

in their own self-interest. Dividends may also curb managers‟ tendency for over investing. In

this way, it is speculated that dividends serve to reduce conflict of interests between

managers and shareholders. As dividend payment reduces the overinvestment problem and

15

They suggested that, generally, in countries such as the US, UK, and Australia where common law is

implemented, investors have better protection than in countries governed by civil law such as France, Germany,

and Japan.

73

agency costs, it may have a positive impact on stock price, which is in turn is a critical

determinant of any mergers and acquisition particularly in the banking sector. The overall

impact of the dividend hypothesis on M&A necessitated the rationale for this research.

3.12 Summary

The literature on dividend policy has produced a large body of theoretical and empirical

research, especially following the publication of the dividend irrelevance hypothesis of M&M

(1961). No general consensus has yet emerged after several decades of investigation, and

scholars can often disagree even about the same empirical evidence. In perfect capital

markets, M&M asserted that the value of a firm is independent of its dividend policy.

However, various market imperfections exist (taxes, transaction costs, information

asymmetry, agency problems, etc) and these market imperfections have provided the basis for

the development of various theories of dividend policy including tax-preference, clientele

effects, signalling, and agency costs.

The chapter has reviewed the evolution of corporate dividend policy. It was noted that

dividend policy has been bound up with the development and history of the corporation itself.

The basic argument and M&M proof of dividend irrelevancy have also been streamlined and

the chapter also explored the main theories that counter the irrelevancy proposition of

dividend. To provide an understanding of dividend policy theories, the chapter attempted to

explain the basic argument for each theory followed by the most important empirical

evidence on testing of these theories. The theoretical and empirical studies discussed in this

chapter have established that a number of possible factors can influence dividend policy.

74

Chapter Four

Analysis of Mergers & Acquisitions

4.0 Introduction

The issue of mergers and acquisitions hereafter refered to as M&A has captured the interest

of many both in the business industries and academics. Several of these studies have focused

on the banking sector due to the strategic importance of the banks in any economy. During

the period 1997 - 2007, several M&A were initiated in many EU countries, many of which

were either domestic or cross border deals. This cahpter therefore reviews the roles of M&A,

EU regulations on bank M&A and the impact of the glaobal economic impasse on banks

mergers.

4.1 The Concept of Mergers and Acquisitions

The term mergers is often used in congruence with acquisitions. It is one of the most popular

terms used in finance and economics. Its divergent nature has earned it a wide interpretation

and definitions. Mergers & Acquisitions ( M&A) play crucial roles in the development and

growth of corporate entities. They provide a substantial source of finance to several

organisations. But some other organisations could percieve M&A as a threat to their

independent existence (Watson & Head, 2005).

Academics often use the terms mergers and acquisitions interchangeably or refer to the

acronym „M&A‟ without actually differentiating between the two events. In theory, the

merger between X and Y occurs when there is amalgamation between both to form another

legal entity Z. This contrasts with acquisition that occurs in any three ways: when X buys

minority stake in Y( less than 50%); when X buys majority share in Y (above 50.01%) of Y

and when X buys all the shares in Y(100%) (see, e.g. Pasiouras et al, 2005).

Mergers can be defined as a reorganisation of assets into a new organisation. They often

75

involve companies with similar size. This is in order to avoid the dominance of one company

by the other. This feature differentiates M&A from takeover which is often associated with

companies of different sizes. The bigger company often called the „bidder‟which offers to

acquire the smaller or target company‟s shares. This can be done by paying cash, offering

shares in their company (equity purchase), or both. In practice though, the line between

mergers and takeover is narrowed as one of the merging companies is often bigger than the

other.

Cybo-ottone & Murgia (2000) defined merger and acquisition ( M&A) from the economic

perspective. They explained that M&A can be considered as an event that involved

companies along the profit function through a change in size, scope and distance from the

efficient frontier. This definition suggests that profit is a veritable product of M&A. They

also suggest that evidence on the the cross section estimates of profit and cost functions may

provide prooof of any profitability impacts resulting in a merger deal.The above definition

further highlights that M&A is the acquisition of a stake of 50% without control, if the

transfer is between affiliated parties or between different parties.

4.1.1 Types of bank mergers

Mergers can be classified into horizontal, vertical and conglomerate . Horizontal mergers take

place between competitors, while vertical mergers occur between companies that are on

buyer and seller relationship. Conglomerate merger is an amalgam of firms of different lines

of operation.

Horizontal mergers occurs when a company takes over another fom the same industry and at

the stage of production process. Thus, the horizontal integration involves the combination of

two competitors (Gaughan, 2002). Horizontal mergers are more difficult to implement when

the market is oligopolistic (only a few firms on the market) and therefore can result to

monopolistic power. In the UK, the hostile bid in 2001 by Lloyds TSB for Abbey National

plc was prevented as this merger would have given too much power in the market to Lloyds

TSB. It was therefore perceived as against public interest. The primary motives of a

horizontal merger is to achieve economies of scale and market power.

76

Vertical mergers occur when organisations acquire another with the desire to achieve scope.

This happens where the target is in the same industry as the acquirer but operating at a

different stage of the production chain. Depamphilis (2003) explains that vertical integration

can better be demonstarted with the use of corporate value chain. Organisations that do not

own operations in each major segment of value chain may choose to backward integrate by

acquiring a supplier or to forward integrate by acquiring a distributor. Thus, vertical

integration reduces costs of research, contracting, payment collection, advertsing,

communication, and co-ordination of production. The synergy achieved in a vertical

integration may produce size and cost efficiency for the integrated companies.

Finally, conglomerate mergers occurs where the target is in an activity apparently unrelated

to the acquirer although the same activities such as marketing may be related. Gaughan

(2002) opines that a conglomerate merger occurs when the companies are not competitors

and do not have a buyer–seller relationship. Some conglomerate mergers are motivated by

risk reduction through diversification and to improve efficiency.

4.1.2 Determinants of Bank mergers and Acquisitions

Bank mergers do not happen by accidents. They are often carried out with caution and care.

This is because of the complexities involved in a merger deal. Berger & Humphrey (1992)

emphasized that the stock market data produce some evidence on the impact of a merger deal

on the shareholders of the companies. Therefore to ascertain the efficacy of a merger deal as

at the time of it‟s annoucement can be established by splitting the sample according to the

size, scope, geophaphy, and the legal nature of the merger. A positive impact of the merger

can be explained by an increase in the efficency, profitability or the share price following the

merger.

Scope : The merger deal is undertaken when there is existence of economies of scope. This

can be achieved by comparing the value of the shareholders creation effect of similar (eg.

bank to bank) as opposed to cross product deals.

Geography: The geographical dimension of the deal is very relevant. Rhoades (1993)

explained that a merger deal occurrs when a more efficient firm acquires a less efficient firm.

77

This means that a bank merger may be motivated to eliminate the duplication of activities.

Such mergers often occur when there is a considerable overlap in the market.

Economies of Scale: Some evidence of economies of scale effects can also be observed if the

value creation effects are larger than the size of the deal. The size of the deal could provide a

rough measure of market concentration following the deal. This yardstick is usually included

in some merger guidelines.

Economies of scale occurs because the scale of operations is larger after the M&A.

Economies of Scale are most likely to arise in horizontal acquisitions but may also surface in

vertical acquisitions. They are common in areas such as production, distribution, marketing,

management and finance (Watson & Head, 2005).

Synergy: This happens when the assets of the operations of the companies involved in the

deal complement each other. Thus the combined outpout after the merger is greater than the

sum of their individual outputs.

Legal: The legal nature of the M&A gives a better understanding on the deal. If the deal is

prompted by desire of acquisition, it will follow from a desire to preserve the identity of the

target by limiting intervention to the injection of capital. Whereas a desired increase in cost

efficiency will be more likely to be related to a centralised mode of organisation like a

merger. In reality, the legal distinction between the merger and acquisition is blurred. Some

acquired firms are later merged with bidder companies as a result of the friendly nature of

many acquistions, especially in the banking sector which is highly regulated (James &

Houston, 1996).

Comparing the bidders and target banks: Another determinant feature of M&A is the

assessment of the pre-merger performance of the banks. If the preformance has been poor, it

will be evident that the merger is prompted by injecting efficiency in the management of the

bank. Watson and Head (2005) infer that a declining share price will attract potential bidders

who are convinced that they can manage the company more efficiently.

78

4.1.3 Reasons for Bank Mergers and Acquisitions

There are often justifications for banks to undertake M&A. The reasons can be classified

under two broad categories: economic and financial justifications. Watson & Head (2005)

enlist the economic reasons for M&A as entry to new markets, provision of critical mass,

means of providing growth, and increase in market power and share. The financial reasons

range from financial synergy, target undervaluation, tax considerations and increase in

earnings per share.

Banks may wish to expand even beyond their domestic territory. Such cross border expansion

can be facilitated by mergers. This could include the expansion into new geographical and

business areas of the bank especially when the internal growth is not too efficient.

The critical mass or size may be essential in the credibility of the bank. Mergers are

sometimes critical in enhancing the needed attractiveness of the bank to attract such fanfare

from the public. Given the veritable nature of their research and developments; marketing,

investments, merging companies can put their resources together to muscle the critical mass

and size essential to provide sufficient funds to finance such activities (Altunbas & Marques,

2008).

When a bank gets to its elastic limit, the chances of internal growth becomes uncertain.

Mergers and acquisitions provides a good opportunity for a bank to strategize its growth

focus. M&A also increases the chances of acquiring market share and may strengthen the

banks monopolistic power especially in horizontal mergers and acquistions. However, the

vertical acquisition tightens the grip on raw materials or the distribution.

Financial synergy occurs if the banks reduce their cost of capital as a result of M&A. This

benefit can be achieved when there is an increased size following an acquisition since such

acquisition can drastically reduce the interest rate on new debts. Rad & Beek (1999) add that

synergies in a merger deal is more realistic when the costs of the merged firm are lower than

79

the costs of the two merging banks which again is a direct benefit of economies of scale.

Another important aspect of merger is its use in eliminating ineffective management team at

least cost.

Another interesting reason that facilities a merger is the availability of information about the

target bank or the information hypothesis (Hawawini & Swary,1990). This can be obtained

when there is a correspondent-respondent relationship between the banks, particularly in

cross border mergers. Banks may have private information about another where there exists

such correspondent-respondent relationship. Such privately held information can be utilised

by the bidder banks to evaluate a merger attempt.

4.1.4 Bank Mergers and Shareholders Wealth

The past two decades have experienced an enormous increase in the number of mergers

involving banks and other financial institutions particularly in Europe. The cumulative money

value of the merger activities from 1997 to 2007, for instance is believed to be in upwards of

96 trillion dollars (Reuters, 2009). The US dominated the world merger activities but the

trend has also highly infiltrated the Eurpean financial market.

According to European Economic Research (1996), the number of M&A activities in the

Eurpoean countries increases by over 50-90 deals per year . This is at an estimated average of

15 mergers per year. The financial yield resulting from such deals amounts to about $77.9

billion while in the US, it amounts to $193.6 billion within 1991 and 1996. Thus, till date the

US bank mergers have been the focus of merger activities among most researchers due to the

volumious amount of information available.

There have been numerous studies relating to the impact of M&A on the profits of banks

and shareholders waelth. Some of the studies have indicated both negative and positive

effects on the both the bidder and target banks and their shareholders. Eckbo (1983) found

that when a bank merger is announced, the share price of the bank tends to rise. This is either

because the increase of market concentration improves profits of the larger players. This

result contrasts with some others undertaken especially in the US such as Cornett and

Tehradnian (1992), Houston and Ryngaert (1994); Siems (1996) which found a decline in the

80

share prices of the banks after M&A.

The result obtained by Cybo-Ottone and Murgia (2000) suggest that merger increases the

share value of the banks. This finding is notable as it streamlines a higher positive result in

mergers involving banks and other financial institutions such as insurance firms. The returns

on such mergers were statistically significant than in other bank-to-bank mergers.

Additionally, the study shows that in the year following the deals‟ annoucement, there is

higher and significant abnormal returns .

The study of Rhoades (1993) on the domestic bank mergers finds some weak evidence that

bidder bank underperforms in the market more than the targets whereas the opposite is

claimed in cross border deals. The financial institutions have a clear pattern of significant

under performance in the years prior to the announcement of their deal. The study also

amplified the difference in the results of a merger as different from an acquisition. It

compared the post merger performance of the target banks that have been acquired with

targets that have been merged, and finds a very small and non significant under performance

of merger targets as compared with acquisitions targets and with merger bidders. This finding

is consistent with that of Cybo-Ottone and Murgia (2000) in that there is weak evidence that

merger deals have a more disciplinary appeal than acquistion.

The argument of Houston & Ryngaert (1994) was based on their study from a sample of US

mergers and could not establish any significant input on the value of the shareholders wealth

particularly the acquiring bank. The study concludes that the shareholders wealth is in effect

transferred from the acquiring to the target banks. However, a positive correlation between

the value created as a result of the merger and some other factors were established; such

factors as the acquiring bank‟s operating performance, the nature of the operating

environment, and the mode of financing the deal.

Although the issue of how much wealth is generated or reduced by merger has remained very

contentious as well as controversial, most researchers agree that there is some impact on the

shareholders or the banks. Some studies attribute the wealth changes to a greater efficiency

resulting from the merger; and the pooling together of resources (Pillof, 1996; Berger, 2000;

Hughes et al 1999 ). While others such as Beitel et al (2004) suggest that most mergers result

81

in excess returns to the shareholders as a result of combination of both size, profitability of

the banks and the efficiency factor. However, they concluded that the shareholders wealth

will be higher when the target bank has a poor stock performance and the acquirer bank is

involved in few merger deals.

However, Kane (2000) adds that such wealth generation resulting from merger is likely to be

feasible if there is a potential for increase in market concentration and where the target is a

large depository bank and both are located in the same area. This undermines any argument

in support of cross border M&A. Mergers involving large banks are favorably disposed as

their share price is likely to improve.

However, Becher (2000) offers an insightful finding regarding merger activities in the EU in

the 1990s. His findings suggest that wealth creation by banks mergers was at its peak over

the period, and this is supported by the large increase in bank mergers and deregulations

during the period. The number of banks mergers increased by over 2.15% over the past

decade. The study indicated that merger in the banking sector created more synergies and

benefits to merging banks through time. Using a 36 day event window, the study concluded

that the wealth created by the merger was positively significant to the targets as well as the

returns to the bidders. The overall effect is significantly positive and larger than any other in

the 1980s. However, over an 11- day event, the result of the study produced a positive return

to the target but negative to the bidders.

The analysis captured the opinion of some previous studies which suggest a longer event

window for the targets when calculating the abnormal returns. The study also reaffirmed the

positive impact of deregulation on the banking sector and upheld the importance of synergies

in M&A transactions.

There is another argument about the value created by mergers. According to the principal –

agent theory, there is an assertion that the interest of the managers do not often represent

those of the shareholders in the consideration of a merger deals. This is particulary common

where the managers renumerations are based on growth or size. Thus, managers will be more

inclined to engage in mergers irrespective of the benefits or doubts surrounding such

acquisitions. Roll (1988) confirms that in some circumstances, the target banks could be

82

overpayed in a display of pride of managing a mega bank at the expense of the shareholders.

This latter arguement is often refered to as the „hybris hypothesis‟.

Bank mergers are also figured to be tools for reducing the operating expenses and thus

increasing net profits and efficiency. However, Berger and Humphrey (1992) demostrated

that mergers of banks with totals assets worth over $100 million have an insignificant effect

on the operating costs of the merged banks. In fact, some studies have failed to find any

collaboration between mergers and economies of scale in the banking sector. Bogus banks are

not often the most profitable firms (Boyd & Graham,1991). This tends to suggest that large

and merged banks do not necessarily create good shareholders wealth or have as good

efficiency as may be expected.

4.1.5 Bank Mergers and Growth in Bank Formation

The overall impact of mergers in the banking industry can be assessed by the number of

functional banks existing after a merger wave. Arguments have been advanced purporting

that consolidation results from deregulatory changes which shrink the number of banks. The

European Central Bank report (2004) maintains that number of EU banks fell by over 23%

following the mergers that occurred between 1971-2003. The incidence of M&A is often

found to be the aftermath of deregulations in the sector, technological changes, increase

competition in the industry, introduction of the euro. As a result, inefficient banks have been

absorbed by the more efficient ones.

However, Cabral et al (2002) noted that the majority of mergers occurred among the small

domestic institutions and among banks operating in the EU countries. Banking groups have

been criticised for concentrating their merger activities within the domestic banks. This ploy,

Campa & Hernando (2006) argue, is to consolidate their positions within their countries of

operation rather than targeting their strategy to other foreign EU countries where uncertainty

is greater and survival as a result of competition is made difficult.

A significant factor in merger activity is the geographical diversification which results from

the merger. There is also the improved competitiveness and the ability of the banks to provide

services and values to the consumers by the cross selling of their products.

83

There are different opinions on the reasons for the rise in bank mergers. Two studies

conducted have divergent views about the impact of bank mergers on the formation of banks.

Such divergent findings raise questions about the correlation of bank mergers and bank

growth or formation. Keeton (2000) opined that the mergers that encourage new bank

formation are those that involve the takeover of small banks by the large banks or those

involving the takeover of local banks by distant banks. This can be adduced by the

disappearance of small banks.

Results on bank merger activities in the 1990s indicate that mergers are strong factors in the

growth of banks. Markets with high record of merger activities have higher growth of bank

formations especially those involving the takeover of small banks by the big banks or the

local banks by the distant banks. This can be attributed to high profitability, increased

population and economic growth rates and low market concentration (Amel and Liang,

1997).

The issue of bank mergers and growth came into focus with the upsurge in banks M&A

which coincided with the growth of banks in the 1990s. This incidental increase in merger

activity provides a lead to the conclusion by some economic analysts that merger is a crucial

index of such bank growth and formations. This conclusion was however challenged by the

Seeling & Critchfield (1999) which found that heavy merger activity and new bank

formations are not correlated. The incidence could both have increased in response to other

factors such as high profits, strong economic growth etc. The study opined that instead,

mergers in fact act as deterrent to bank formations.

The emergence of the above findings led to further research on the issue by Berger et al

(1999). They confirmed the popular opinion that merger activity actually increases bank

formation. The study did not only consider similar factors examined by Seelin & Critchfield

(1999), it also used data on mergers and new bank charters across different markets and

controlling for other factors that could have encouraged mergers and new bank formations

within the 1990s.

Others have held bank M&A contributes to the significant reduction in the number of banks.

84

Goddard et al (2007) argued that bank consolidation had reduced the number of banks in the

majority of the EU15 countries particularly within 1984 - 2004. This assertion was earlier

highlighted by Walkner & Raes (2005). In evidence drawn from the Europen Central Bank

(2004), they examined the percentage decline in the number of domestic banks between 1997

and 2004 which was attributed largely to merger formation in the various member states of

EU. Table 4.1 presents their data.

Table 4.1 Trend in the Number of Banks Among EU Countries

Source: ECB(2004) and extracted from Walkner & Raes(2005)

As can be noted in the table, there is a decrease in the number of bank branches in most of

85

the countries. In Belgium, the number declined by more than 32% while in Germany and the

Netherlands, the number fell down by 25.1% and 46% respectively. The UK and

Luxembourg also recorded a fall of 13.2% and 15.4% in the number of local branches while

in Sweden, it fell by 18.2% . However, an increase in the number of bank branches rose to

more than 31% in Greece, 19% in Italy, 14% in Portugal, 4% in Spain and 1.3% in France.

This appears to be an inconsistent trend as majority of the EU countries show a decrease in

the number of bank formations.

The argument for and against the effect of merger activity can be understood by considering

the economic impact on banks. Mergers can discourage bank charters by producing

organisations with enough local market power to discourage entry. On the otherhand, mergers

can also encourage bank charters by creating a niche in service to small customers who are

more concerned about their personal service. Such service will hardly be a priority of the big

merged banks.

4.2 Studies on European Domestic & Crossborder Bank Mergers

There has been a surge in banks mergers within and across the European countries over the

last decade. Records show that domestic mergers have maintained steady growth in relation

to cross border or international mergers. However, between 1985 and 2001 only about one

fifth of all merger transactions were cross border. The banking sector accounted for only

13% of the such deals (Focarelli & Pozzolo, 2001). The rare occurence of cross border

mergers has been attributed to several factors such as inefficiency, cultural issues, adverse

regulation, communication barriers and distance which have contributed to its limited success

in such deals. Goddard et al (2007) confirmed that cross border bank mergers particularly

those involving large banks have been quite unpopular within the European countries.

Domestic banks tend to outperform international banks in terms of their efficiency

particularly in developed economies. Amihud et al (2002) thus assert that cross border

mergers in such respect give a picture of weak competition in the market.

Most studies on cross border deals have concentrated on inefficiency, nature of the target and

acquirer banks and regulatory framework as crucial factors in its limited success (Vander-

86

Vennet, 1998; Berger et al 2000). However, Buch & DeLong (2004) added that information

costs remain a considerable determinant of international merger deals. This is in consonance

with Berger et al (2001) which add that cultural differences, supervisory structures, common

currency have inadvertently inhibited cross border mergers in Europe.

Buch & DeLong (2004) applied the gravity model of foreign investment decision to diagnose

the rationale behind cross border mergers. The model has been previous applied to foreign

trade-in-goods between two countries, which is a factor of the market size and the geographic

distance between the countries. The distance between the countries is considered to be a

proxy for transportation costs. The expense on cost created by distance has been upheld as a

critical issue in the growth of cross border M&A. Other studies (Wei & Wu, 2001; Portes &

Rey, 1999) concur with their assertion.

Another key variable in the determinant factors of cross border M&A has been the

deregulatory environment. Countries with effective privatisation programmes have attracted

foreign investors particularly in the fragile banking sector. Most European countries with

privatised banking sector have been points of interest for merger deals. Bonin & Abel (2000)

found that privatisation has been a key influence in the high market shares of foreign banks in

the Eastern Europe.

Conversely, economies with more stringent information requirements and efficient

supervisory system will likely win the confidence of foreign interests. They often have laid

down disclosure regulatory requirements which give vital information of their operation and

performance. Thus banks in countries with such high regulatory standard will be more

attractive for international acquirers in cross border merger deals. However, Buch &

DeLong( 2004) caution that such high regulatory standard, if too high may be a disincentive

and could impede foreign interest from considering cross border merger deals.

Ongena and Penas (2008) investigated the determinants of the bondholders‟ wealth effects of

the acquirers in domestic and cross border European banks mergers in the periods of 1998 -

2002 and concluded that the abnormal returns to domestic bondholders is higher than those of

the cross border banks. The study also indicated that the banks‟ bondsholders experience

abnormal returns of up to 5% higher than those participating in the cross border merger when

87

the acquirers country has a strict banking policy and regulations.

Other studies have corroborated the above findings. Penas & Unal (2004) find that

bondholders of both domestic and cross border banks in the US mergers realise significant

positive risk and maturity adjusted returns around the merger announcement month. After

controlling for a variety of merger and bank characteristics, the bondholders returns show

increase in the asset size as a result of the mergers. In the same vein, Choi, Hasan & Francis

(2007) examined the stock market reactions to merger announcements involving cross border

acquirers. They find the impact to be related to cross country difference in the deposit

insurance and investors protection but with little relevance to the supervisory regimes. The

study however failed to state any direct impact of the domestic and cross border mergers on

the bondholders‟ wealth of the banks.

Aw and Chatterjee (2004) find that UK domestic acquirers achieve better performance than

the UK cross border acquirers , particularly those acquiring US targets. The UK domestic

acquirers show positive but non significant abnormal returns in the six months following the

merger announcement. The cross border EU banks acquisitions have significant negative

returns. This argument is supported by various studies such as Danbolt (1995), Conn (2003)

and Black et al (2003) which examined cross border acquisitions of quoted targets. Conn

(2003) finds that of the 15 cross border merger studies reviewed, the resonant conclusion is

the zero or negative abnormal returns which applies to both UK and EU acquirers. These

abnormal zero and negative returns are also found in the domestic acquisitions (Sudarsanma

& Mahate, 2003).

More recent studies have expressed diverse opinions and views on the profitability, efficiency

and wealth generation of the cross border and domestic bank acquirers. Mangold & Lippok

(2008) compared the shareholder wealth created by cross border with those of domestic

acquirers among German firms. Their findings indicate that while domestic acquirers create

wealth, cross border banks destroy shareholders wealth. The wealth destruction by the foreign

acquirers in explained by the difficulties in post merger integration due to cultural

differences, legal and economic obstacles. The study also adds that mergers involving non-

EU targets result in positive abnormal returns while EU targets results in significant negative

cumulative abnormal returns to the shareholders of the acquirers.

88

The above assertion is supported by the findings of Lensink & Maslennikova (2008) who

established that domestic bank acquirers create shareholder value while cross borders

acquirers do not. The finding is based on the geographic and product market diversification

hypothesis. They applied the excess demand and the barrier to entry theories (Brewer et al,

2000) to explain how geographic deregulations might affect the cross border announcements.

While the excess demand theory posits that as geographical restrictions to mergers are

eliminated, the number of bidders competiting for a given target will invariably increase.

This will in turn increase the purchase price paid by the acquirer banks. In most cases, the

target might be overpriced with the subsequent long-run effect of underperformance in

relation to the abnormal returns of the merged bank. The study, however, used a window of

20 days which might not capture the essential post merger events impacting on the acquirer

banks performance.

The barrier to entry theory advance the idea that the prices paid by cross border acquirers will

be lower as the barriers to the merger deals are removed. The argument is that the barriers

often provide a bulwark to the targets which inevitably attracts the abnormal returns common

with targets. Thus, eliminating the barriers to entry increases the chances of replacement of

the target banks and lowers the purchase or merger price premium. This ensures that the

target is not overvalued and the acquirers are less likely to have poor abnormal returns. The

theory thus upholds that barrier to entry encourages abnormal returns to the acquirers.

Delong (2001) studied the market reaction to a geographical diversified bank mergers and

finds that the US stock market reacts positively to domestic acquirers as against cross border

acquirers in a merger announcement. The reaction can be explained by the potentials for cost

savings which is greater when the banks are exposed to similar economic situations. This

argument is also supported by the findings of Cybo-Ottone and Mugria (2000) who find that

domestic acquisitions create more abnormal returns to their shareholders while cross border

acquirers show a positive but not significant returns.

However, not all cross border versus domestic merger studies support the assertion that

domestic acquirers create more value to the shareholders than the cross border acquirers. In

fact, Bertrand and Zitouna (2008) find the EU cross border and domestic bank acquirers

89

differ significantly. The study debugs any fears cast on cross border acquirers as they help

enhance both the shareholders value and the domestic country‟s economy. The efficiency

gains are found to be stronger in the cross border than domestic mergers and acquisitions.

The difference in both acquirers can be explained by the mechanism of the European

economic integrations.

For cross border acquirers to create more wealth to the shareholders, they have to be

distinctive in size and operations. Anand, Capron and Will (2005) add that only cross border

acquirers with multinational scope are capable of creating more wealth for the shareholders.

The implication is that only cross border bank acquirers that are conglomerates are capable of

inflating the returns of the shareholders. In the same vein, Ecko and Thornburn (2000) and

Gregory and McCorriston (2004) studies both private and quoted cross border M&A but find

no trace of significant negative returns.

However, some studies have found no difference in the returns of the both the cross border

and domestic acquirers. Lowsinski, Schiereck and Thomas (2004) remark that any difference

in the wealth effect of the domestic and cross border acquirers may be due to market

segmentations in imperfectly competitive markets. In a ten day window to evaluate the effect

of the cross border acquisitions on Swiss shareholder wealth, the domestic deals recorded

higher abnormal returns than the cross border bank shareholders. On the merger

announcement date, the cross border acquirers indicated positive returns while the domestic

acquirers showed negative returns. However, the empirical result of the study showed no

significant difference between the cumulative abnormal returns of the both acquirers. The

study has a shortcoming of being limited to Swiss bank acquirers.

The above assertion is corroborated by Hernando, Nieto and Wall (2008) who find that both

the cross border and domestic acquirers show no much difference in their characteristics. The

cross border coefficients variables of EU bank acquirers are positive and statistically

significant than the domestic acquirers, while the abnormal returns to the shareholders are

high for the cross border acquirers. However, the study finds that cross border acquirers are

less likely to increase their market shares in the target domestic market. Cross border

acquirers are often attracted by the possibility of having a higher rent which might be readily

available in more concentrated markets.

90

Domestic acquirers are less likely to be profitable in more concentrated markets than their

cross border counterparts. Hernando et al (2008) also reported that domestic banks are likely

to acquire large banks. This can be explained by the theory that large banks are more

beneficial in achieving both diversification and enhancing market penetration. There is also a

suspicion that supervisory authorities tend to create easier platform for the acquisition of

large banks by the domestic acquirers.

Conn et al (2005) find cross border acquirers have lower announcement but long-run returns

than domestic acquirers. However, in the short run, the study find domestic acquirers have

negative abnormal returns while the cross border show zero and negative postmerger returns.

While cultural difference is significant in the long run merger performance of the domestic

acquirers, such a factor has no significance in the cross border acquisitions. Also taxes,

accounting standards, the legal system and exchange rate volatility have no significance in

the returns and performance of the acquirers. Conn et al (2005) maintain that cultural

difference remains an odd for the cross border acquirers especially when integration and

acculturation are vital factors in the process. The greater the cultural gap, the worse the

problem of the success of the deal for cross border acquirers as opposed to the domestic.

4.2.1 Studies on the Performance of Domestic & Cross Border Bank Acquirers

Essentially, performance studies in mergers and acquisitions can be categorised into 7 broad

areas: integration process performance (Bresman et al., 1999; Capron, 1999; Datta, 1991;

Larsson & Finkelstein, 1999; Shanley & Carrea,1992); overall acquisition performance

(Bruton et al., 1994; Hayward, 2002; Humburg & Bucerius, 2006; Puranam et al., 2006;

accounting performance (Palich et al., 2000; Zollo & Singh, 2004; Feea & Thomas, 2004,

Goddard, Molyneux & Wilson, 2004a; DeLong & DeYoung, 2007); long-term financial

performance (Carow et al., 2004; Harrison et al., 2005; Palich et al., 2000); short-term

performance (Moeller et al., 2004; Pangarkar, 2004; Seth et al., 2002; Shahrur, 2005; Walker,

2005; Capron & Pistre, 2002; Carow et al., 2004); acquisition survival (Bergh, 2001;

Vermeulen & Barkema, 1996; Pennings et al., 1994) and innovation performance (Ahuja &

Katila, 2001; Kapoor & Lim ,2005; Hitt et al., 1991, 1996 & 1998).

91

Examination of some operational performance studies indicate mixed results. For instance,

Cornett & Tehranian (1992) studied the performance of 30 US bank acquirers and find a

correlation between stock market reactions and changes in the financial performance. This

implies that the market reacts commensurately with the level of the performance of the banks.

This can be seen when banks announce their year profits and losses. Pillof (1996)

investigated 48 mergers over a period of 10 years to establish the correlation between the

abnormal returns and the financial performance of the banks and find no significant results.

Which implies that the market speculations at the time of the merger announcements failed in

predicting the actual results.

In most merger performance studies, the acquirer banks are often found to have negative or

no significant returns. Past studies such as Baradwaj, Dubofsky & Fraser (1992) find a

negative returns for the acquirers in a study that involved 108 mergers for a period of 6 years.

Cornett & Tehranian (1996) investigated mergers with minimum deals of $100 million over

a period of 5 years and find that the majority of the 30 mergers studied indicate that the

acquirer banks recorded negative returns to their shareholders. Even on mergers deals with

values exceeding $500 million (or the megamerger deals) acquirers showed negative

abnormal returns to the acquirer banks shreholders (Siems, 1996). Hart & Apilado (2002)

developed a GARCH market model to sustain an argument that cross border acquisitions

though have positive stock returns, their post preformance are poor.

Some others have used larger samples to arrive at the same conclusion that the acquirer bank

shareholders earn negative returns in comparison with the domestic acquirers.

Subrahmanyam, Rangan & Rosenstein (1997) used 225 merger deals to support their

arguement that acquirer shareholders have negative returns. While Houston & Ryngaert

(1994) and Madura & Wiant (1994) examined samples of 153 and 152 of both cross border

and domestic acquirers and find the returns negative.

Studies on the evaluation of merger financial performance have produced varied results.

While some studies have judged performance based on the reaction around the announcement

date of the merger, others have hinged their judgement on the returns accruable to the

shareholders. However, Olson & Pagano (1998) had indicated that any meaningful evaluation

92

of a merger should be based on the long term of the acquirer‟s stock price perfomance rather

than on a short term basis. Such long-term performance will include the returns to

shareholders, dividend payout, and the share prices subsequent to the merger deal.

This argument concides with Houston & Ryngaert (1994) that a veritable measure of merger

performance is not solely the profit but should incorporate bulk elements of the returns to the

assets. This is a cautious measure since accounting data may not provide the correct state of

the merged bank in the short run especially in terms of costs and returns. Therefore, it is

understandable why an efficient market may provide a better understanding of the market

rather than the accounting data. Their figures are mostly unbiased as against the financial

accounting results that are often subjected to management influence.

It is also claimed that the past financial performance of banks involved in mergers could play

crucial role in the determination of the expected returns to shareholders in a merger.

However, Berger & Humphrey( 1982) argued that a deal involving an efficient bank

acquiring an inefficient bank may result in reduction in operating costs through the

elimination of X-inefficiencies. The contention is that the acquiring bank should be more

profitable than the acquired and the revaluation of the share prices of the both banks should

be related to the difference in their profits.

However, the posture of the bidders as more profitable than the target does not explain the

abnormal returns from such deals. This corroborates the studies of Houston & Ryngaert

(1994) and Morck et al (1990) who find a significantly higher abnormal return in bank

mergers involving more profitable bidder banks. Thus suggesting that bidder banks with

strong records of high profit profile will produce a merger deal which will be more wealth

creative to the shareholders than those with weak profit background.

Another aspect of evaluating the performance of M&A is the effect from the welfare of bank

customers. An expected overall impact of banks would be an improvement in the lending

rates and access to credit facilities to customers. Study by Sapienza (2002) shows that when a

merger involves domestic banks operating in the same geographic area, borrowers tend to

benefit from good lending rate particularly if the acquired bank has small market shares. The

finding of the research tend to support the argument that small borrowers benefit more in

93

domestic M&A because the large cross border acquirer banks tend to do away with the small

borrowers.

This position differs from the claim made by Scott & Dunkelberg (2003), that M&A have no

noticeable effect on the ability of small firms to obtain loans but could trigger increase in

poor service quality by small firms. However when banks merge, the acquiring firm

determines the banks lending policy. If the acquirer does not have a significant lending

culture, this could impact on the small entrepreneurs who may find it difficult benefitting

from the M&A. Peek & Rosengren (1997) argued that a significant number of small banks

who are involved in M&A usually have a business loan portfolio. The impact of this is the

eventual promotion of small business financing. Thus the impact of M&A is often judged on

the amount of support given to small business lending.

Kosmidou et al (2004) applied multivariate analysis using financial variables to compare

domestic and cross border banks performance in the UK. The study reveals that on average,

domestic banks have higher operating performance. Specifically, the domestic banks record

high performance on the rate of their return on equity, loans to short term funding ratio and

net interest revenue to total assets. Although, this study is not focused on M&A, it does

however project the „home advantage hypothesis‟ which assumes that domestic banks are

more efficient than foreign banks in their operational performance due to their knowledge of

the local environment.

The length of the event window could be very critical in the measurement of the success and

failure of the acquirers. Short and long windows may inevitably suggest different

performance results for the same or different acquirers. Zollo and Meier (2008) confirm that

the short term acquisition and firm performance are often measured by the abnormal returns

to the acquirers over a short window around the announcement period. The difference

between the short term acquisition performance and the long term lies in the market

expectations and actual result of the merger. While the former is focused on the premerger

collective financial market expectations and events, the latter window is provides some

knowledge of what has actually occurred during the merger process. Infact, they represent

the ex ante expectations and the ex post factor performances of the merging units and the

merged entities.

94

Thus, the present study leans towards using long term window of -/+30 days (60days) in

order to capture, not only the short market reaction of the merger announcement, but only

using financial ratios of the merged entities to evaluate the longterm operational performance.

Some studies (Harrison et al. 2005; Zollo & Meier, 2008; Barber & Lyon, 1997) have

confirmed that long windows are far more realiable than the short windows. This argument

is hinged on the widespread evidence of market imperfection to the announcement of M&A

which can be reduced by a long window. In fact, short windows are deemed to give

misleading results since it captures dominant cognitive performance. This research therefore

takes further steps in strengthening the long window argument; first it uses a long window in

estimating the abnormal returns of the acquirer banks and applies some key financial ratios

in evaluating the performance of the banks over a period of 11 years. Such long evaluation

period is uncommon in performance studies.

4.3 Mergers & Acquisitions and Bank Efficiency

Walker & Raes (2005) deliberated on the different levels of efficiencies associated with bank

M&A. These include, but are not limited to, operating profit and overall efficiencies. Finance

literatures are diverse with views indicating lack of visible improvements in the operating

efficiency of merged banks. In a study by Rhoades (1994), two approaches were applied to

explore the operations of consolidated banks using 39 empirical works. The operating

performance approach and the event studies were simultaneously applied.

The operating approach takes into consideration the financial performance of banks after

mergers while the event studies observed the share price reactions subsequent to and after the

merger announcements. The approaches failed to establish any significant improvement both

in the banks financial performance and rise in the share price of the consolidated banks. This

also corroborated the findings of Pillof & Santomero (1997) that M&A do not contribute to

improvement both in share value and profit of banks.

The above scenario has generated concerns as to possible reasons why efficiency cannot be

improved with merger. A number of findings have been advanced by various scholars.

95

Vander-Vennet (2002) associates this to a precipitation of a takeover of a poorly performing

bank by a relatively efficient bank. In such deals, there may be improvement in the short term

operating profit but not in the operational efficiency. The differences in the countries tax and

legal systems are possible players in the unrealisation of the potential synergies particularly

in cross border mergers.

Another argument presented by Hughes et al (2003) is that operating efficiency may be

lacking especially when the management of the consolidated bank seeks to build empire

rather than consolidate on its existing valuable resources. An increased in acquired assets

may increase or decrease operating efficiency depending on the attitude and posture of

management. Inordinate ambition of expansion; acquisition of more banks and other similar

agency objectives may deter efficiency. Managers are more inclined to achieve efficiency if

they streamline their resources in achieving organisational objectives rather than engaging in

other empire building course.

In the US, the variation in accounting rules have been posited as a cause for the discrepancy

in establishing operating efficiency in the M&A in the banking sector. Two accounting

principles: the purchasing and pooling of interest accountings, which are in use in the US

often produce conflicting results in their methods of treating M&A. The purchasing

accounting method triggers the rise in the consolidated banks depreciation and amortisation

expenses. This is due to the treatment of goodwill (i.e the purchasing price of the acquisition,

less the book value of the target banks assets and equity) which is amortised and expensed as

they are intangible assets.

The pooling of interest method equates the assets of the consolidated bank with the sum of

the two merging banks. Kwan & Wilcox (1999) opined that the use of the purchasing

accounting method will inevitably produce a negative result on the reported expenses. Both

methods are however recognised by the US GAAP ( Generally Accepted Accounting

Principles) .

96

4.4 M&A & The Corporate Governance of Banks

The resounding opinions on the conflict of interest of managers and directors who ordinarily

should be the stewards of banks and their shareholders. Their actions often indicate a selfish

motive and thus portray conflicts of interests. Most mergers and acquisitions have suffered

basically because the „agents‟ have allowed their interests go ahead of those of the „bank

owners‟. This is particularly so with the acquiring banks, where various studies indicate that

their shareholders suffer deficiencies in their returns.

Zalan and Lewis (2005) examined the reasons behind bank acquirers CEOs engaging in

mergers and acquisitions, even when there is strong evidence that the result might not

generate returns for the acquirers‟ shareholders. Their finding indicates that the managers and

their M&A promoters have developed a non efficiency based posture. The relationships

among the CEOs, the banks promoters and the board of directors gives a strong argument for

the preference of mergers rather than the agency relationship with the shareholders. In other

words, the managers would prefer to please their promoters by engaging in unprofitable deals

as mergers rather than considering the consequence on their shareholders returns.

The above argument is further strengethened by the study of Certo et al., (2008) who found

that directors are more concerned with the size of their pay package rather than the

shareholders interests. They affirmed that when directors are faced with decisions including

M&A that can lead to an increase in their pay, they would opt for their own interest. By so

doing, they become agents rather than the stewards who ought to protect the interests bank

owners. Aggarwal & Samwick (2003) also confirmed that managers diversify their firms in

response to the changes in their personal benefits rather than reducing the exposure to risk of

their firms.

However, Coakley and Iliopoulou (2006) argued that although pay packages might be an

element in the consideration of mergers and acquisitions by CEOs of bidder firms, their

rewards however are usually tied to the success of the deal. Managerial power are more of

97

interest than agency theory on the their executive compensation. The US CEO acquirers are

found to be more compensated than their EU counterparts on the successful execution of a

merger and acquisition.

As the agency problems become more diverse in organizations, the duties of the board in

monitoring the activities of the management take broader dimension particularly when the

company‟s performance is poor. A poor performing bank may be an indication of the

inefficiency of the inside directors and could warrant the replacement of the inside directors

with outsiders or even the removal of the CEO.

Although the inside directors are perceived to have an understanding of the business

operations, the independent or outside directors are professionals and can more easily achieve

the supervisory role, reduce the occurrence of conflicts among the top management and also

prevent the misuse of the company‟s resources, thus improving performance (Hayes, 2005).

This implies that they exert enormous influence on the company and act to protect the

shareholders interest. Their position as outsiders gives them the moral leverage to question

any indiscreet acts of the top management.

Outside directors exude some independence on the CEO. They could act independently as

they are not in the control of the top management. Unlike their inside directors who are

almost handpicked by the CEO; Weisbach (1988) indicates that independently dominated

board are likely to dismiss a poor performing CEO. They are portrayed as doing more jobs to

protect the interest of the shareholders, including decisions on engaging in mergers. Boards

with majority of outside directors are perceived to positively affect the share price of the

firm16

.

The presence of directors generally is to mitigate between the shareholders and the top

management of the firm. Their duty is to curtail the agency problem which arises when

management goes outside the interest of the shareholders. Fama & Jensen (1983) outlines the

board‟s functions as including the ratification of the management decisions and the

monitoring of the management performance. They are by law empowered to hire and dismiss

16

A study by Rosenstein & Wyatt (1990), shows that there is a rise in the stock price of the firm at the

announcement of an appointment of an outside director to the board. This strongly suggests that shareholders

have high confidence in the outside directors.

98

managers and to determine their compensations. Such authority leaves the directors with

enough powers to oversee the management.

When a high proportion of the firm‟s shares are owned by the members of the management,

there is a tendency that they will be more committed to benefit the shareholders and to raise

the firm‟s share value. The presence of outside directors will thus be inconsequential as the

agency problem is almost invisible. However, Hermalin & Weisbach (1991) opined that even

when majority shares are held by top management, it does not signal an absence of agency

problem.

Such scenario points the direction of family ownership of the firm which are known for

putting the family interest at the expense of the other minority shareholders. Top management

or board positions could be reserved for family members, foundations can be created to

honour family members using the firm‟s resources. Such an act even calls for a more

stringent action, to protect the other shareholders.

An important criterion for determining the extent and necessity of involving greater

percentage of outside directors in the board composition of an organisation is the length of

service of the top management and the CEO. If the management has been in office for a long

time, it may be an indication of their good performance; otherwise they would have been

fired by the board or even taken over. But this may not be leverage for the CEO to stay

indefinite. Vancil (1987) states that the general consensus is for CEOs to serve less than 10

years in order not to loose focus and will reflect a penchant of dominance by the CEO. This

may adversely affect the performance of the firm.

The composition of the board is definitely a concern of the CEO17

, whose performance is

under the purview of the board. It will therefore be an understatement to expect that the CEO

to show serious interest on who constitutes the board. While the choice of the inside directors

may be influenced by the CEO, the outside are independently made. They are usually

engaged on the basis of their professional experience and directorship in other reputable

firms. Borokhovich et al (1996) agree that the inside directors increase the value of their

human capital by strengthening their reputations as decision control experts. But the inside

17

Studies have shown that CEOs are well involved in the appointment of the directors. Mace (1986) opined that

the CEO will be interested in persons who will be less insurrogative to his decisions.

99

directors however are more concerned with maintaining their position in the firm and would

not be involved in confrontations with the CEO.

However, Borokhovich et al (1996) maintain that the influence of the CEO in the

appointment of the directors may hamper the independence of the directors and will thus

affect their keenness in protecting the shareholders‟ interest. This argument suggests that the

directors may after all, not represent the shareholders interest. CEOs that pursue interest

outside those of the shareholders would be interested in influencing the appointment of

directors who will support their decisions.

Whether the independent directors actually represent the shareholders interest has been an

unending debate. Some studies (Gilson, 1990; Rosesnstein & Wyatt, 1990; Hermalin &

Weisbach, 1988) affirm that outside directors represent the shareholders interests.

Independent directors are defined as those whose relationship with the bank is the fact they

are members of it board of directors. Cotter et at (1997) explain that an independent board is

one with above 50% majority of members as non executive directors. They opine that due to

the high confidence in their judgment, boards with majority of independent directors usually

have fewer takeover bids on grounds of performance.

Infact, the shareholders confidence in the quality of the independent directors is shown by the

study of Byrd and Hickman (1992) which shows that the share price of the firm increases

when the board is controlled by the independent directors and also when an outsider has been

named as member of the board. The abnormal returns on the date of takeover announcement

are higher when the decision is perceived to be made by the independent directors‟ dominated

board rather than the inside.

Various prior studies have focused on the presence of independent directors, corporate

performance and ownership structure particularly in the banking sector (Weir, 1997 and

Adams & Mehran, 2003). Most studies on independent directorship have focused on firm

performance (Zhilan et al, 2005; Adams, R. & Mehren, 2003, Weir, 1997) and Rosestein &

Wyatt (1990) and Block (1999) focused on US firms.

The independent directors in any organization are very fundamental in the formulation of

important policies. Board ownership especially outside or independent directors play key

roles in effective corporate governance. Brook et al (2000) observed that independent

100

directors bear some losses during mergers as they often loose their place in a merger. Their

equity ownership is thus considered as having greater marginal impact than those of their

counterpart insider directors. They are considered to be very influential in the decision of the

board to sell or merge with other banks. Their composition in the board is thus significant.

In addition to the oversight functions, the independent directors are charged with substantial

powers to ensure that corporate decision makings maximize shareholders wealth. Some

studies (Wyatt, 1990; Byrd & Hickman, 1992 and Cotter et al, 1997) have confirmed that

boards with majority of independent directors soar very high in their performance, share

value, high abnormal returns during takeover bids and generally improves the shareholders

returns.

Several studies have been carried out on board composition and performance particularly in

the banking sector. Adams & Mehren (2003) compared the corporate banking structure in the

banks with those of manufacturing and other sectors. Their findings show significant

difference in the board size, percentage of the outside directors, frequency of boarding

meetings, percentage of the CEO equity holdings and other such variables which were higher

in the banking sector. The CEO equity holding, the stock option pay, salary plus bonuses in

the banks are higher than in other sectors.

The tenet of the agency theory spurs from the argument that the management and decision

making is divorced from the ownership of the organization. Where the decision making is

delegated to the agent by the principal, and if there is a discrepancy in the interests of the two

parties, agency problems will arise. This implies that corporate governance, in addition to

protecting the interest of the shareholders in the running of the bank, also minimizes the

agency costs and reduces the breadth of managerial discretions (Vafeas, 1999).

Fama & Jensen (1983) found that the composition of the board in publicly quoted companies

provide such avenue for monitoring the excesses by the management and ensuring that

shareholders interest prevails rather than those of the management of the bank. While the

difference between the UK and the US board structure in the number of directors may be

blank, studies suggest that the difference lies in the composition of the executive and non

executive directors. In the UK, the board is dominated by the executive or inside directors

while in the US, the non executive directors are usually in the majority (Main & Johnston,

1993).

101

The directors are usually appointed based on their experience, skill and expertise. The non

executive directors have dual function of monitoring the management and also exercising an

oversight role over the board activities. As outsiders, they have a more neutral stand in their

decisions. Fama (1980) opines that efficiency and values are also measured by the number of

directorship they hold. Reasonably number of directorship in other firms signals the worth

and experience of the directors. However, an average of three directorships is considered

reasonable by Cadbury (1992) in order to reduce the amount of time committed to other

functions.

The major contribution of board structure to the performance of the organization is the

prevention of fraudulent activities and the enhancement of smoothness of the firm‟s

operations. Firms that have good corporate governance structure often have impressive

operating performance. Such firms also provide useful information to the investors and their

creditors to enhance their operations and reduce the information asymmetry (Demsetz &

Lehn, 1985).

The independence of the directors can be impaired or achieved by structure of the

directorship of the board members. Where there are interlocking directors18

in board of bank,

the decision of the interlocking directors are likely to be impaired especially if the CEO of the

bank is also an independent director in banks where they act as CEO.

If the independent directors hold substantial percentage of the bank‟s shares, they interest to

protect the shareholders become more intense as they are also concerned with increasing the

shareholders wealth. Morck, Schleifer & Vishny (1989) support the argument that firms

performance increases when the independent directors hold majority of the shares while

Shivdasani (1993) opines that takeover bids are high where the independent directors own

small proportion of the banks shares.

18

An interlocking directorship occurs when a director acts as an executive director or CEO in a firm and at the

same time, as an independent director where the CEO is an independent director. Okazaki, Sawada, &

Yokoyama (2005) assert that such directorship is capable of slowing opposition to the CEO irrational policies

and therefore questions the extent of shareholders protection offered by such board.

102

4.5 European Union Merger Regulations & Control

All mergers and acquisitions in the EU are required to comply with the laid down rules and

regulations issued by the Union or its affiliate institutions. The European Commission

(hereby referred to as the EC), is the executive arm of the European Union responsible for

proposing legislation, implementing rules and regulations of the Union. The essence of these

regulations is to ensure transparency and smooth M&A deals and to intervene where

necessary when the rules are breached.

There have been a series of periodic merger control reforms by the EC some of which have

been very radical and with a far reaching impact on execution of bank mergers among

member countries. These reforms and regulations are aimed at streamlining the guidelines for

mergers involving competing firms, best practice guidelines on the conduct of merger and

investigations in line with some jurisdictional implications.

EU merger activities came under stringent control on September 21, 1990 to foster credibility

in M&A competition among EU countries. The overall aim of the regulation is to protect the

interest of consumers by encouraging the consistent existence of competition but at the same

time, discouraging monopolistic tendencies occassioned by mergers. The regulation is

founded on four pivotal issues:

1) The exclusive right of the Regulatory Commision to review merger activities within

the EU states. Such activities will also include acquisition and joint venture

businesses.

2) The Regulation makes it compulsory for any merger deals to be notified to the

Commission. The Commission has set standards regarding the size for mergers that it

considers to be subject to mandatory notification to the Commission.

3) The consistent application of competitive based criteria for M&A. The commmission

considers third parties in cases of dispute to make oral presentations and encourages

interested bodies to play active roles in EU merger regulations.

4) The provision of a legal framework to govern any merger activities. It also enjoys

exclusive jurisdiction on merger regulations though its decisions are subject to

judicial review by the member state courts.

103

Levy (2003) asserts that since the institutionalisation of the merger regulation, the

commission has given unconditional approval for over 1800 applications of merger related

deals; 50 others have been have been turned down; another 50 have been referred for

consideration by the EU member countries. While 95 merger transactions have recieved

conditional approval during the first phase of the Commission, during phase two of the

Commission, 20 mergers deals have been given unconditional approval; 18 others prohibited

with three of such prohibited overruled by a member country court and about 80 applications

to the Commission have been withdrawn .

The commission has worked towards ensuring that its enforcement standard is adhered to by

member countries. To this end, the Commission issued a Draft Horizontal Mergers Notice to

explain how mergers should be analysed and identifying various factors that may impede a

fair competitive atmosphere. The adoption of this Draft has other wider implications and

benefits to the EU states. At least, it creates a platform to achieve standard through an

assessment of M&A transactions.

Cross border acquisitions have been hampered by different regulatory frameworks across

countries. La Porta et al (2000) posit that investor protection is highest in English common

law colonies as well as in Scandinavian, German and French civil law territories. Such legal

environments influences cross border acquisitions. Efficient cross border acquisitions will

take place where an acquirer from a high investor protection country acquires a target from a

low investors protection country. Information asymetric may also slow the performance of

cross border acquirers due to retsrictions imposed by the regulatory regimes (Moeller and

Schlingemann, 2002).

104

4.5.1 History of EC Merger Control Legislation

The development of the EC merger control, or control of concentrations, has been solidified

with the gradual creation of a European single market. European competition policy has

historically revolved around articles 81 & 82 of the EC Treaty19

. Since the EC Treaty is silent

on the control of mergers, the supervision of concentrations was in the early stages of the

Union restricted to the Commission‟s control of compliance with article 82. In the precedent

set by the Continental Can case, when the European Court of Justice (ECJ) held that an

undertaking abuses its dominant position if it strengthens its position in such a way that the

degree of dominance reached substantially fetters competition, the way was clear for the EC

to rule on M&A.

The limited supervision implied by the treaty provisions was problematic since it restricted

the control of mergers to those that were made between dominant undertakings. It is

interesting to note, that the Commission as early as the 1960s had realized this shortcoming

(Dauz, 2002). However, it would take another 30 years until a legal instrument dealing solely

with concentrations was agreed upon by the EU member countries. The reason for this delay

might partly be explained by the political importance that is linked to an instrument through

which mergers between undertakings may be restricted or prohibited. Massive lay-offs, new

investments or withdrawn investments are in many cases very tangible results of mergers.

The governments of the EU were reluctant to surrender these powers to the Union.

As in many other fields of Community law, it was a judgement of the ECJ that put an end to

the power squabble. In the BAT-Reynolds merger dispute, the Court conferred competence on

the Commission to supervise mergers under article 81 of the EC Treaty (in addition to article

82) which created an atmosphere conducive for the adoption of the Merger control

Regulation

It was the Regulation 4064 of 1989 that introduced the basic principles upon which the

current merger control as contained in European Commission Merger Regulation (ECMR)

139 of 2004 is based. Regulation 4064 of 1989 clearly spelt out the Commission‟s exclusive

19

The EC Treaty governs the historical formation of the European Union. Article 81 of EC Treaty prohibits any

agreements between undertakings while Article 82 deals with the abuse of dominant position in a merger

transaction.

105

jurisdiction to clear or prohibit concentrations with a Community dimension. It introduced

the “one-stop-shop” mandatory notification procedure for the concentrations that fell within

its jurisdiction as well as market-oriented, competition-based criteria. The Commission was

empowered to impose fines and competence to restore competition by ordering the undoing

of a prohibited merger transaction. The strict deadlines for decision-making provided for

legal certainty for the parties to the concentration. Finally, decisions of the Commission are

subject to scrutiny by the Community courts.

These principles manifest themselves also in the current merger control legislation

particularly Regulation 139 of the EC Merger Regulations (ECMR). However, developments

in case-law and the limited resources of the Commission has led to some reforms, among

which the most important include the possibility of referral to national authorities of

concentrations with a community dimension under article 9 of the ECMR and a reformed

substantive test which allows for the intervention against more types of anti-competitive

mergers.

Nonetheless, the system of ex ante administrative control of mergers and the Commission‟s

wide powers of investigation and enforcement established by the Merger Regulation is ever

still the foundation of the Community‟s merger control. The introduction of the merger

regulations meant that the Commission‟s authority to supervise mergers was explicitly laid

down in Community legislation. Similarly, the establishment of an administrative system of

merger control has important implications on the role of the Community courts and the

standard of judicial review they ought to apply (Murray, 2004).

If BAT-Reynolds merger dispute was the starting point for a more extensive control of

concentrations, the merger Regulation 4064 of 1989 provided the Commission with a

foundation to build the extended control on. However, as conventional in EC law, the

performance of the Community Courts finally shape and determine the application and

interpretation of legislation.

106

4.5.2 European Union Bank Reforms and Mergers Regulations Amendments

The EU merger regulation came into effect on the 21 September, 1990 through the enactment

of the EC which has the robust powers of enforcing Articles 81 and 82 of the EC Treaty. The

Treaty was enacted to prevent the perceived unfair competition and abuse of the strong

market forces in any mergers acquisitions deals. The Articles 82 highlights what constitutes

abuse as when there is a limit placed on the production, markets or technical developments to

the prejudice of the consumers by an unfair trading of a member state. This also includes any

perceived unfairness in the fixing of prices in a business transaction.

The Article also explains that any condition that could place any party in an advantage

condition, which at the same time disadvantages another party or places the party in a more

competitive position, is considered „unfair‟ (Harris, 2007).

The approval or rejection of mergers activities within the EU is usually the responsibility of

the EC. Where a merger has been rejected, there is an internal procedure of obtaining a

remedy through the European Court of First Instance (referred herein as CFI).

107

The CFI makes an independent investigation and reviews decisions either by giving approval

to the merger or rejecting its execution.20

Any further remedy sought is usually referred to the

European Court of Justice (ECJ)21

which serves as the court of final instance for all merger

issues (Motta, 2000).

The underlying principle applied by the EC in considering approval for banks mergers is

usually based on the Commission‟s basic litmus tests: the extent of the transparency of the

deal; availability of deterrent mechanism in place to prevent any unfair conduct by parties

and reactions of the undertakings which do not participate in the coordination, such as the

current or future competitors and the customers are required to avoid acts that can hamper the

deal22

.

20

Several merger reviews have been undertaken by the CFI over the years. Some of these decisions have set a

precedent in the merger rulings in EU countries. Their jurisdiction is not only limited to bank M&A but includes

all other forms of consolidations. On July 13, 2006, the CFI made a landmark judgement by putting a stop to the

merger of Sony –BMG, which would have been the second largest music company in the globe. Some mergers

that have been cleared by the EC have also been challenged in the CFI. Also in 2003, the EC objected to the

General Electric and Honeywell merger even when it has been cleared by the US Department of Justice. The

companies appealed against the decision of the EC asking for the removal of the prohibition but the CFI rejected

the appeal. (See the General Electric V. Commission).

21 Levy (2003) narrates that since the Merger Regulation came into force, the EC has rendered more than 2,200

decisions, of which over 1,800 (86%) had approved notified concentrations unconditionally; around 50 (3 %)

had found the Merger Regulation to be inapplicable; around 50 (2 %) had referred concentrations notified under

the Merger Regulation in whole or in part to the government of EU countries; around 95 (5 %) had approved

transactions subject to undertakings given at the end of the initial investigative period;11 around 20 (1 %) had

approved transactions unconditionally during phase II; around 60 (3 %) had approved concentrations subject to

undertakings given at the end of phase II; and had rendered prohibition decisions with respect to 18 (1 %)

transactions,12 three of which were subsequently overturned by the Community courts on appeal. More than 80

operations had been withdrawn, of which around 20 (1 %) were withdrawn following the opening of in-depth

investigations, in many instances to avoid prohibition decisions. Thus, around 2% cent of all transactions

notified under the Merger Regulation have been either prohibited or abandoned in the course of in-depth

investigations. The Commission‟s challenge rate” is broadly comparable to those of other major jurisdictions.

22 Motta, Polo and Vasconcelos (2007) document detail merger remedies undertaken by the EC over the past

years. Some of the remedies are either of structural and non structural in nature. For instance, in 1997, 188

disputed merger cases were submitted to the Commission; 235 in 1998; 292 in 1999; 345 in 2000; 335 in 2001;

279 in 2002; 249 in 2004; 313 in 2005 and 403 in 2007.

108

In the US however, the procedure for approving or rejecting a merger deal seems to have

clearer guidelines than in the EU. The law requires the merging banks to obtain clearance

from either the Department of Justice (DOJ) or the Federal Trade Commission (FTC).

Unlike the EC system where third or interested parties are allowed to appeal against the

Commission‟s decision, the US system does not make provisions for such. Infact, neither the

DOJ nor the FTC gives aggrieved parties the opportunity to know any reasons for their

decisions, as such decisions are not published. Harris (2007) points out that the while US are

more precautious in handling outside merger interest through their regulations, the EU allows

more freedom from third parties as long as sufficient interests have been proven.

Several reforms have been undertaken by the Economic Commission to overhaul the merger

process in the EU. Byowitz, et al (2002) outlines some major areas of the EU Merger

Control Regulation. Specifically, the reform includes the following areas substantive;

procedural; jurisdictional and administrative reforms. The substantial reform is based on the

concept of the dominance criteria used in assessing the impact and market powers of mergers.

The EC has upheld the principle of the dominance test rather than the substantial lessening of

competition peculiar to the US system. The current reforms define dominance to include

some oligopolies that raise competition issues which tend to differentiate its policy from the

US. The merger reform gives a guideline that helps to identify grey areas in which mergers

can be considered as anticompetitive. These include: the establishment of the paramount

market position which arises when the combined market share is more than 50% or when

there are reasonable grounds to believe that despite having a lower percentage share, a party

will still exert a stronger market power. Motta, Polo & Vasconcelos (2007) also highlight

some remedies put in place by the EC to address any forms of anticompetitive practice or any

merger that will create a dominant position as a result of effective competition would be

greatly affected. They argue that the use of dominance test rather than the US application of

substantial lessening of competition criteria is more useful in addressing any ill competitive

tendency.

Another guideline is in the area of elimination of constraints on the ability of suppliers to

unilaterally alter prices which is a form of non- collusive oligopoly. There are also reforms

against any increased risk of coordination among competitors in the market which is a form

of collusive oligopoly. The EC guideline is to give priority to efficiency in the consideration

of merger analysis. The EC merger reforms consider efficiency as a vital ground for merger

109

initiation. The essence of the argument is that the consumers benefit more where the system

is efficiently developed and are unlikely to create monopolistic tendencies. The Commission

also proves mergers on the basis of imminent failure of the firm when the target would have

likely been forced out of the market and where there is no suspicion of any anticompetitive

acquirers. If the execution of a merger will help to sustain the survival of the target bank

assets, the Commission would be obliged to grant the merger on the basis of efficiency23

.

The procedural reforms include the striking off of the requirement that the parties in a merger

file intention within a week of from the conclusion of a binding agreement between the

merging banks. The reform adopts the US style where an ex ante merger notification and also

requires the merging parties to make a submission notifying its intents before the execution of

a legal agreement. The procedural reforms can also extends the time periods to a further 30

working days for the completion of the merger review where the parties have proposed

remedies to address any competitive concerns raised. This can also be further extended to

additional 15 days in complex cases.

In addition, EC reforms increased the amount of both the fines and penalties of the parties

where they are found to have made any misleading submissions or any such issues raised

during the Commission‟s inquiries. The Commission is also empowered to seal the premises

of the offending parties during an investigation (Byowitz et al, 2002).

23 The efficiency clause has not been without controversies. Christiansen (2005) and Padilla (2004) argue that

the introduction of the efficiencies defence is a reaction to the criticism of the old Merger Regulation. It was

claimed that efficiencies had been falsely used as an argument against the merging parties and that welfare-

enhancing mergers had therefore been prevented. In the sense of the "error cost approach" these would be

„Type II errors‟. However, the empirical evidence for this claim is fairly weak. In their econometric study Neven

and Roller (2002) detect only a few „Type II errors‟. There is no reference either to concrete cases apart from the

General Electric/Honeywell merger and this case is questionable. The prohibition was recently upheld by the

Court of First Instance (CFI), which would appear to argue against a false decision by the EC. Experience with

the efficiencies defence in the US also suggests that this new rule has little relevance in practice. So far there has

been no publicly available evidence of an approval decision based on efficiencies (Colley 2004). Hence, no

significant reduction of errors is therefore to be expected either from the introduction of the efficiencies defence

in EU merger control. Moreover, the concrete conditions are so strictly formulated in the Guidelines that they

cannot be fulfilled in practice (Baker, 2002). The massive problems associated with the application of the

efficiencies defence also need to be considered. In particular, there are information asymmetries in the

authority's disfavour and fundamental knowledge problems.

110

The jurisdictional reforms focus on simplifying the system of referring transactions from the

Economic Commission to the EU member countries which may be affected by the merger24

.

The reform therefore proposes a pre-notification system which allows the merging banks to

seek referral of a merger transaction to any EU country involved or affected by the merger.

The reform also permits the process of acquiring exclusive jurisdiction over any merger deals

involving cross border parties.

Issues included in the administrative reforms include: the creation of office of the Chief

Competition Economist; the appointment of a peer review panel to undertake investigations,

outside interests in a merger conflict. The reform also includes regular meetings with interest

groups in event of conflict in the merger with a view of resolving the issue.

4.5.3 The EU Merger Regulation and Political interference

Since the introduction of the European Union merger control, analysts have spotted several

fundamental institutional flaws in operation and procedures of seeking redress in the

European Commission. Drauz (2004) argues that the responsibility for final decision-making

lies with a primarily political body and its affiliate bodies (the CFI and the ECJ) whose

members are potentially exposed to influence from firms and from governments of member

countries). For example, firms can deliberately seek to influence antitrust proceedings at the

expense of their competitors (Baumol & Ordover 1985).

One possibility is the prevention of mergers between competitors which could lead to

competition-relevant efficiency benefits. This constitutes a form of rent-seeking which, if

24 According to the Economic Commission Competition Policy Annual Report (2007), the number of merger

cases notified to the Commission reached an all-time high of 402, a rise of more than 12% compared to the 356

transactions notified in 2006. In the last quarter of the year the number of notifications fell both in relation to the

previous quarters and the last quarter of 2006. In total the Commission adopted 396 final Decisions in 2007, of

which 368 were cleared in the first phase without conditions. Of these unconditional first-phase clearances 238

(or 65%) were adopted under the simplified procedure. A further 18 transactions were cleared in Phase I subject

to conditions. Ten Decisions were adopted after in-depth Phase II investigations. Five of these were cleared

without conditions, while in four case; the clearances were subject to conditions.

111

successful, results in Type II errors25

in the sense of the "error cost approach". On the other

hand, political influence can be exerted, especially by national governments, aimed at

securing approval for certain mergers despite reservations about the effects on competition.

Under the banner of "industrial policy", Stevenson & Fillippi (2004) opine that there is a

strong interest in building "champions" for the global markets or securing special treatment

for sensitive industries. However, by and large, there are no sound theoretical arguments to

justify this. Moreover, the empirical experience is mostly negative. Consequently, approvals

granted in exceptional cases for industry policy reasons are false decisions of Type I

(Christiansen & Kerber 2005). Hence, in both cases, there is the risk of welfare losses, which

are highly relevant to the discussion of the "more economic approach".

During the initial years of the introduction of the EU merger control, Christiansen (2006)

argued that there were a number of questionable decisions in which influence of this kind

played a role. For instance, in Germany during the Boeing/McDonnell Douglas merger

dispute, it was believed that the Airbus consortium exerted influence on the Commission

(Boeder & Dorman 2000). But there also allegations of massive intervention by the member

state government and in some extreme cases, the members of the Commission were divided

among themselves (Coate & Kleit, 1998).

A sensible institutional solution to this problem would be, in the first place, to create an

independent antitrust authority at the European level which would have sole responsibility for

protecting competition and could build up an appropriate reputation. However, proposals to

this effect met with massive resistance and have receded more and more into the background.

In the present reform discussion this fundamental institutional aspect is receiving little or no

attention.

25

There basically two form of empirical errors; the Type 1 Error and the Type II Error, which are similar to the

EC decisions on merger cases. The Type Error I which relates to wrong approval of a merger (false positive)

while the Type II Error relates to an unjustified prohibitions or denial of a good merger deal by the Commission.

(False negative). The Improved decision-making quality would mean a reduction in the frequency of the two

types of error which can arise. In both cases, the potentially achievable level of economic welfare is not

attained. In the first case, this results in direct welfare losses while, in the second, potential efficiencies are not

realised. The overview shows the scenarios that are possible. According to the "error cost approach" there would

be an improvement in decision-making quality if the new criteria lead to a reduction of errors compared with

practice before the reform (Christiansen, 2006).

112

Admittedly, political factors have been found to play a lesser role in recent years in merger

resolutions by the EC (Pons & Sautter 2004). But this does not suggest by any means that it is

no longer an issue. Rather, there are two further points which need to be considered. The

increased focus on economic analysis can be interpreted as an attempt on the part of the

supporters of a purely competition-oriented approach within the Competition Directorate

General to shield themselves from attempts to exert external political influence. The greater

complexity of the economic argumentation before the reform has doubtlessly been a

contributing factor (Levy 2003). Thus, the more economic approach oriented would lend to a

more a logical refinement of a strategy which has already being pursued.

These measures of reforms are geared at giving a transparent and fully compatible rationale

in merger deals independent of any political-economic aspects of merger control.

Consideration for industrial policy or other non-competitive factors would be in no way

compatible. Perhaps in order to reduce political influence, there should be great measure of

independence of the competition authority; this includes the stronger orientation of merger

control to more general rules so as to reduce the room for discretionary decisions and thus the

exposure to influence. This could also quash the contention that in its more recent decisions,

as in the GE/Honeywell merger for instance, the European Commission is allowing itself to

be influenced by an underlying anti-American sentiment – and thus again by non-competitive

factors (Murray 2004).

Secondly, the fact that political intervention has been successfully pushed back in recent

years does not imply that the problem has been resolved once and for all. Rather, it has to be

assumed that the inclination towards anti-competitive intervention at the political level will

continue to exist, at least latently. This was clearly indicated by the initiative on the part of

France and Germany for a pan-European industry policy while on the subject of Germany,

there was a special ministerial powers exercised in the E.ON and Ruhrgas merger dispute or

the heated debate over Springer-Verlag's acquisition of Pro Sieben Sat.1 (Lingus et al 2004;

Anon, 2005).

So, for this reason, too, there is still a need to give thought to institutional safeguards to shield

the EU merger control process from (industry policy-related) political influence. At the same

113

time, the increasing orientation towards case-by-case analysis within the framework of the

"more economic approach" also creates new possibilities for discretionary decisions. Given

this ambiguity the incentives for firms and politicians to exert influence could be increased

again (Baumol & Ordover 1985). That this aspect has gone largely unnoticed to date points

once more to an overly narrow focus of the discussions on the new approach.

4.5.4 Some Principal Provisions of the EU Merger Regulations

Levy (2003) classifies the powers of the European Commission on Merger Regulations

(ECMR) under four basic areas namely: the exclusive competence of the Commission to

review concentrations of Community dimension; the mandatory notification of such

concentrations; the consistent application of market-oriented as well as competition-based

criteria; and the provision of legal certainty through timely decision making. Infact, he noted

the following eight cardinal provisions of the Regulation:

1). The Merger Regulation applies to concentrations or changes in control of the process. The

concept of a concentration includes mergers, acquisitions, and the formation of jointly

controlled, autonomous, full-function joint ventures. The concept of control is defined as the

possibility to exercise “decisive influence”.

2). All concentrations that meet certain “size” tests are deemed to have Community

dimension and, as such, are subject to mandatory notification under the Merger Regulation,

irrespective of whether they have any effect in the Community. The Commission has

exclusive jurisdiction over such transactions (the “one-stop-shop” principle). Transactions

that fall below these thresholds may be subject to national competition rules. In exceptional

circumstances, an EU member country may request either that the Commission refers a

concentration of Community dimension to its national authority or that the Commission

review a concentration that does not have a Community dimension.

3). Transactions must be notified on a prescribed form according to mandated time periods, a

waiting period must be observed before notifiable transactions can be put into effect, and the

Commission must render a decision no later than five months (90 working days) following

notification. Fines may be imposed for failure to notify, late notifications, or the provision of

114

incorrect or misleading information. Where reportable transactions have been implemented

prior to having received approval, the Commission may take remedial action.

4.). The Merger Regulation provides considerable scope for third parties to comment on

notified concentrations, including the right to be heard orally. The Commission encourages

customers, competitors, suppliers, and other interested parties to play a active role in EU

merger control.

5). The substantive test under the Merger Regulation is whether a transaction creates or

strengthens a dominant position as a result of which effective competition is significantly

impeded in the common market. The Commission‟s appraisal under the Merger Regulation

has two main elements: (i) definition of the relevant market; and (ii) competitive assessment

of the transaction. The Commission generally focuses first on unilateral exercises of market

power and then on whether a transaction creates or strengthens a position of collective or

oligopolistic dominance. Horizontal mergers (i.e. those involving firms active in the same

markets) have accounts for the large majority of challenged transactions, although the

Commission has also examined (and, on occasion, prohibited) mergers that have had vertical

or conglomerate effects.

6). The Commission is not empowered to exempt or authorise, on public interest or other

grounds, concentrations that are considered incompatible with the common market. The

Commission may peg its approval of transactions on undertakings or commitments offered

during the initial investigative periods.

5) An appraisal under Article 81, which prohibits anti-competitive agreements, may also be

warranted under the Merger Regulation where a full-function joint venture gives rise to spill-

over effects between its parent companies. Non-full-function joint ventures fall outside the

Merger Regulation and may be subject to Articles 81 or 82, which prohibits abusive conduct

by dominant companies, as well as national competition rules. Any decisions of the

Commission are subject to judicial review by the Community courts.

115

4.5.5 The US and EU Merger Regulations

There is a sharp constrast in the pattern of merger regulation and the economic environment

of the EU and the US. Rad & Beek (1999) noted that the EU member states are very

heterogenous. The differences in culture, legal and economic systems are substantially

diverse than the US. But this also has some impacts on any merger transactions. It

strengthens the diversifications caused by the loose integration in economies of member

nations.

There is also the fear that economies of scale will be difficult to exploit in an EU cross

merger. The difference in the legal operations of the domestic country implies that in an

occassion of a merger, the merged bank will need to readjust to the demands of the local

legislations. However the US offers a less stringent measure in the legal requirement of

laying off staff in event of a merger. In the EU, the labor laws are tight and almost difficult to

allow banks to shade off their staff strength as a costs saving measure. The protections under

the EU regulations are thus stronger.

There is also the issue of limitations on the investment ventures businesses such as banks,

can be allowed to explore. While the US seems restrictive, the EU is more liberal on the

range of activities banks can undertake. For examples, banks are not allowed to undertake

commercial and investment banking. But the EU regulations allows banks to undertake either

or more such as insurance. This inadvertently gives EU banks a measure of economies of

scale and scope in comparison with the US.

4.6 Bank M&A & the Current Economic Crisis

The current global economic crunch has adversely affected the trend of mergers and

acquisitions particularly in the European market. The economic impasse has hit very hard on

the bank sector leading to cover 16 large failed, unsuccessful and cancelled mergers and

acquisitions in the sector (Reuters, 2009). In genral terms, the crunch has cramped the

private equity model as there have been smaller and fewer deals. Young (2008) observes that

owing to the economic crunch, the number of the M&A deals reduced by half in 2008 and the

biggest value of M&A in the chemical industry for instance, being valued at maximum of one

116

hundred and fifity (€150) million euros. Banks want much more security before giving credits

or loans. This has slowed down the ability of the banks to lend.

However, the effect of the credit crunch and the global economic meltdown has not been so

badly felt in all European countries. The German banking system has shown remarkable

resilience in the face of the global financial crisis. Dammers (2008) observes that despite the

collapse of the IKB Industriebank and SachsenLB mergers which was among the earliest

signals of the severity of the credit crunch, the Germany banking system has remained quite

steady and resilient.

But despite its steadfastness, the German banking industry has been faced with its share of

the financial storm leading to the governemnt to introduce broad measures such as bailouts

involving liquidity debt guarantees and recapitalisation of the banks in addition to the

introduction of the Financial Markets Stabilisation Funds to streamline successful takeovers.

Consolidation has been argued as the panacea to the melt down in the banking sector. Some

German banks are also mapping out strategic models to neutralise the financial crisis such as

geographical diversification into eastern and central European countries, scaling down public

sector lending and increasing commercial banking for small businesses by a higher

percentage.

Interestingly, at the start of the credit crunch, some financial analysts had envisaged the

impeding credit squeeze in the global economy which would invariably affect bank

acquisitions. Shearer (2006) outlined that the impending credit crunch facing dealmakers in

the M&A markets in 2006 would have dire consequences for the deals. Tightening of the

credit cycle, reliance on the traditional funding sources such as the leveraged loans were

among the strategic options for the banks.

The increase in the bank interests by most banks in 2006 was a move to tighten lending and

thus signaling a severe financial crisis. Flexibility in the M&A structures such that borrowers

could have sufficient capital to withstand the liquidity reduction was the strategic way to deal

with the credit crunch. The result has been the enormous pressure created on both the

domestic, cross border and opportunistic acquisition deals in the EU banking sector

( Carruthers, 2007).

117

The impact of the global economic crisis on banks M&A has a mixed effect in the US. It has

infact generated a process of creating thriving banks out of many troubled ones; and thus

opened the avenue for big bank mergers. Hage (1991) had alerted that mergers could create a

handful of megabanks with an advantaged position of exploiting cross border acqusitions

and thus rejuvenating the bank sector. However, consolidations alone is not a panacea to the

credit crisis. Both the US and EU have witnessed in recent past bank consolidations leading

to poor performance and returns. However, costs cutting through mergers can be very

helpful. Banks with low overheads earn higher profits and in turn attract more investors

which incraeses the banks liquidity and laons to its customers. Two major issues in the

banking sector are vital for the survival for the credit institution: the reforms in the banks and

its credit risk management system as well as encouragement of survival of smaller banks

who are known for their prudent lending.

4.7 Conclusion

The role of bank M&A in the EU economic community has been reviewed. The chapter also

higlights different factors affecting domestic and cross border acquisitions. The domestic

M&A are more common than cross border deals this is attributable to various factors such as

the EU regulations and competitive advantage. The merit of any M&A is measured on the

shareholder wealth created. Thus it is important especially for bank directors to ascertain that

the aim of any acquisition meets the longterm objectives of the bank.

118

Chapter Five

Research Methodology

5.0 Introduction

Research method involves a systematic and orderly approach applied in the collection and

analysis of data used in a study so that useful information and meaning can be derived from

the data (Jankowicz, 2000). The object of this chapter is to explain the statistical methods and

procedure used in carrying out the empirical analysis, as well the rationale for using such

procedures.

In fact, the study combines the event study approach and regression methods to examine the

impact of the cumulative abnormal returns on dividend policy. Since the study examines

these two very important layers in finance: shareholders returns on mergers and acquisitions

and dividend policy, the research design adopts a combination of regression and event study

methods.

The study is a quantitative research. The design of the M&A aspect of the study follows the

pattern of studies, such as Pillof (1999), Campa & Hernando (2006), Chong et al (2006) and

DeAngelo et al (2006), in the use of an event study methodology with daily share price data

and employing the market model. The dividend policy design used some vital variables

employed by Rozeff (1980) and Olson & Pagano (2005), but in a more constructive format

tailored to amplify the M&A effect. Unlike previous studies, however, this study uses the

standardized abnormal returns in evaluating the shareholders wealth of the cross border and

domestic bank acquirers. The chapter discusses the research design, data sources, sample,

and the data analyses techniques used.

119

5.1 Research Design

5.1.1. The Event Study

The research methodology using the event window is very popular in the field of finance,

insurance and accounting. The technique is particularly used in assessing the effect of market

wide events such as regulations or government legislation, changes arising from M&A,

sudden events in the economy such as recession or shocks in the industry. The methodology

varies depending on the motives of the research.

Rad and Beek (1999) emphasise that the event study method has the advantage of separating

two types of returns: the normal and abnormal returns as a direct consequence of an event; in

this case a merger or acquisition. The technique has been adjourned as the most reliable in

identifying the stock price reaction to specific events such as M&A, and the impact thereon

on the firm. Henderson (1989) highlights among others; three major strengths of event

studies. These include market efficiency, information usefulness and metric explanation

studies.

The market efficiency event studies undertake the investigation of how fast and accurate the

market reacts to new information and development. While the information usefulness

considers the degree to which the returns of the bank is being affected by some piece of

information. The method is also valuable in assessing the professional conducts of managers

and accounting staff and the impact on the reputation of the business (Cross, Davidson &

Thornton, 1988).

There are some identifiable problems associated with the event study method of research.

The complexity and the intrigues of the approach create a strong demand for the justification

of the method. Sometimes, the outcome of events can be so obvious that it paralyses the

efficacy of conducting an event study. Another challenge of this method is the determination

of the event date. This is not the date of the news but the date in which the market‟s reaction

is established. The event window is the date of the occurrence of the event plus or minus the

period it took the market to react to the news (Henderson, 1989).

In any case, the tenet of the method is mainly for empirical business research where

economist and statisticians apply statistical theory to define the best approach to undertake a

120

test, often using simulations. The process involves the following steps as outlined by

Henderson (1989) which were applied in the study:

a) Clear definition of the date when the market would have received the news of the event,

b) Characterization of the returns of the banks prior to the information of the event,

c) Estimation of the difference between the observed returns and the situation without the

event or news; this becomes the abnormal returns,

d) Aggregation of the abnormal returns across the banks and within the time frame under

consideration, and

e) Statistical test to determine the extent of significance of the abnormal returns and the time

span.

The above steps also incorporate some of the 8 listed procedures outlined by Seiler (2004)

which include:

1) Identification of the event date; this is the date on which the event occurred.

2) Definition of the event window

3) Definition of the estimation period

4) Selection of the sample of firms

5) Calculation of the „normal‟ (non-event) returns. These are the returns that would have

occurred in the absence of the event.

6) Calculation of the abnormal returns. These are the actual returns that occurred

because of the event minus the return that would have occurred without the event; the

non-event returns.

7) Calculation of the cumulative abnormal returns (CARs), which is the aggregate of all

the abnormal returns (ARs).

8) Finally, the determination of the statistical significance of the ARs and the CARs.

Each of the above procedures is briefly explained in relation to their applications in this

study.

121

5.1.2 The Identification of the Event Date

The event date is the time the market first heard the news of the mergers and acquisitions.

The news of the M&A may have infiltrated the market either formally or informally, such as

through rumours from insiders, the media or a formal board announcements. The Reuters

(2008) database describes such event dates as the announcement dates of the M&A deal.

The study identifies the announcement day as the date in which the M&A information

reached the market. The dates were obtained from the Reuters database which contains such

data while information regarding the dividend payouts was obtained from the Bankscope

database26

. The market model was used in estimating an Ordinary Least Square (OLS) model

for the periods –130 days to -29 days before applying the coefficients estimates to compute

the abnormal daily returns for the test period (event day -30 to +30).

In each of the events, daily frequency or stock data relating to the sample banks were

obtained from the Yahoo finance. The use of daily stock data is preferred over weekly or

monthly data. Weekly data have the characteristic of infrequency and inability to pinpoint

the exact event time which invariably has the potential of weakening the result of the

empirical test (Seiler, 2004).

5.1.3 Definition of the Event Window

The event window is the length of period preceding the date in which the event (M&A)

occurred. This is usually stated in trading days27

. While there is no consensus window or

period for any study, a more specific period contributes to the strength and significance of the

study (Seiler, 2004 & Well, 2006).

This study used an event window of ± 30days with the event date as day zero. Some studies

such as Becher (2000) have used event windows as short as ± 5days. However, it is important

that the event window should cover the entire effect of the M&A.

26

Different databases are available which have information on the banks and market returns but on various

parameters. For instance, the Reuters database holds weekly returns while the yahoo finance has daily returns on

selected markets or stock indices. Other databases include zephyr, bankalmanac etc.

27 The trading days are the official trading days of the stock market. This usually runs from Monday to Friday

and excludes public holidays and weekends.

122

5.1.4 Estimation Period

The estimation period is used in measuring the expected or normal returns for each bank

during the window. There are four available periods for making the estimates; these are at the

time of the event window, after the event, around the time of the event window and before

the occurrence of the event. In most M&A studies, the estimation period is used with a

disclaimer clause to assure the end users of the research that the period was sufficient enough

to avoid bias in the results (Henderson, 1989).

This study like most other M&A studies28

used the period prior to the M&A to establish the

estimation period. Thus, an estimation period of 101 days (-130 through -29) was used in

determining the abnormal returns for the study. This is illustrated in Fig. 5.1 below. In order

to avoid contamination of the event, care was taken not to allow any overlap of/between the

estimation period and the event dates. Contamination is the overlap of the estimation period

and the event window. The idea of ensuring that the estimation period and the event dates do

not overlap is to ensure that the estimate of the abnormal returns is unbiased; and to show

how the stock would behave in absence of the M&A event (Seiler, 2004).

101 days Event date

--------------------------------------------------0------------

-130 -29 -30 +30

-----Estimation period----- ---Event Window---

Figure 5.1 Graphical representation of the study time frame

28

The most popular estimation technique used in M&A studies use the period before the event (Chong et al,

2006; Cybo-Ottone & Murgia, 2000; Becher, 2000 and Houston & Ryngaert, 2004). It is rare though to have

estimation periods after the event except where the event was so dramatic that it changed the fundamental

relationship between the behavior of the stock and the market.

123

5.1.5 The selection of the acquirer and target banks samples.

Certain criteria were used in obtaining the sample banks in the study. In order to be included

in the sample, both acquirer and the target banks had to meet the following criteria:

i.) The M&A deal must have occurred between January 1, 1997 and December 31, 2007.

ii.) The acquirer bank must have acquired a 10% stake29

or more in the target bank.

iii.) Both the cross border and domestics acquirers banks are commercial banks as classified

by the Reuters database.

iv.) The acquirer banks are member countries of the European Union and have available

financial statements for the periods 1997 to 2007.

v.) All the banks in the sample are listed and publicly traded in the financial market

vi.) Only deals that are classified as completed are included in the list. That means that

pending and cancelled deals are excluded from the list.

vii.) The M&A value must exceed €500 million30

.

viii.) Banks in non EU countries which are acquired are removed in the list of target banks.

The initial total sample of the bidders was 108 banks in the EU. The sample was further

subjected to checks as to determine whether the banks met the set out criteria and 62 banks

made it to the final sample list. Table 5.1 shows the samples from the EU member countries.

29

The 10% is considered as the popular minimum acquisition stake in most finance studies involving M&A and

was applied in the study. See also La Porta, et al (1999), Faccio, et al (2001) and Maury & Pajuste (2002).

30 This €500 million cut off is based on the value of the mega M&A stored by the Reuters database.

124

Table 5.1 List of commercial to commercial banks M&A deals in EU countries 1997 -2007

Year

Country of

Acquirer

Type of

Acquisition

Total per

year

Cross

border Domestic

1997 Italy 0 2 2

1998 Spain 1 0

France 1 0 2

1999 Spain 1 1

Italy 0 1

Germany 1 0

Austria 1 0 5

2000 Spain 2 0

Slovenia 0 1

Italy 1 0

Germany 1 0

United Kingdom 0 1 6

2001 France 3 0

Spain 1 1

Italy 2 0

United Kingdom 1 0 8

2002 Greece 0 1

Belgium 1 0

Germany 1 0

Italy 1 1 5

2003 Germany 1 0

France 1 1

125

Greece 0 1

Italy 0 1 5

2004 France 1 0

Spain 0 1

United Kingdom 2 0 4

2005 United Kingdom 1 0

Austria 1 0

Sweden 1 0

Italy 2 1

Germany 0 1 6

2006 Germany 0 2

Italy 4 1

France 2 0

United Kingdom 1 0

Spain 2 0

Cyprus 1 0

Portugal 1 0 14

2007 Italy 1 2

Greece 1 0

Belgium 0 1 5

TOTALS 42 20 62

126

5.1.6 Calculation of the normal, abnormal and cumulative abnormal returns

The normal returns are the non-event returns that would have occurred in the absence of the

event. The abnormal returns are the actual returns that occurred because of the event less the

returns that would have occurred without the event. The cumulative abnormal returns

(CARs) are the aggregate of all the abnormal returns (ARs).

The abnormal return (AR) is estimated using the market model as:

- - * -----------------------------------------------1

Where

Abnormal return on share j for each day t in the event window

= return on share j for each day t in the event window

= intercept term for share j measured over the estimation period

= slope term for stock j measured over the estimation period

= return on the market m for each day t in the event window

The AR was standardised to cater for the different degree of event impact. This is done by

weighing the abnormal returns by the standard deviation. The purpose of the standardization

is to ensure that each abnormal return has the same variance (Serra, 2002). Thus, by dividing

each firm‟s abnormal residual by the standard deviation over the estimation period, each

residual has an estimated variance of 1 and thus defined by the equation:

= ---------------------------------------------------- 2

Where = SAR for firm j at time t.

= AR for firm j at time t.

127

=

Square root of the variance of the AR for firm j at the time t equals the standard deviation of

the AR for the firm j at the time t.

And the variance is given by the equation:

=

Where:

= variance of the AR for firm j at time t.

= AR for firm j at time I over the estimation period

= number of observed trading day returns for firm j over the estimation period

= return on the market at time t over the event window

= mean return on the market at time t over the estimation period

= return on the market at time t over the estimation period

To determine the significance of the standardized abnormal returns for each day in the event

window, the Z-statistics was used which is expressed as:

128

Z - = ---------------------------------------- 3

where;

= Z- statistics for the each day in the event window

= TSAR for each day in the event window

= number of observed trading day returns for firm j over the estimation period

N = number of banks in the sample

The cumulative abnormal returns (CARs) of the acquirers were obtained by first aggregating

all the abnormal returns in the sample. This follows the suggestion of Campbell et al (1997)

and MacKinlay (1997) that the CARs be aggregated in order to draw the total inferences for

the event under consideration, which could be carried out through time and series. The

cumulative abnormal returns (CARs) were calculated over event period to ascertain the effect

on the period surrounding the announcement of the mergers. For the purpose of the panel set,

the CARs are aggregated across time and bank samples.

5.2 Methods of Measuring Returns

Both the abnormal and normal returns follow a peculiar process in their measurements. The

methods of measuring both returns can be grouped under two broad categories: the economic

and statistical models. While the former involves the application of statistical methods and

assumptions, the latter rely on the reactions of the investors and do not totally lean on

statistical proofs (Campbell et al., 1997). In reality though, it is very typical to combine both

methods in undertaking any empirical work.

A number of models have been developed in the measurement of the firms‟ performance.

These include but not limited to the following: constant mean return model, the market

model, factor model, the market adjusted model etc.

The constant-mean-return model is the simplest model. This is perhaps because of its ability

to give results that are similar to other sophisticated applicable models (Brown & Warner,

129

1985). The lack of the sensitivity to the choice of the model lies on the fact that the variance

of the abnormal returns is frequently not effected by the choice of a more sophisticated

model. The model can be applied when using daily data to measure nominal returns and when

used in measuring monthly data. It can also be used in estimating both excess and normal

returns. The model is expressed as:

Rit = μi + ξit -------------------------------------------------------------------------------------- 4

E (ξit) = 0 and variance (ξit) = σ2

ξi

Where:

Rit = the period t return on security i

μi = the mean return for the assets i and ξit = the time period t disturbance term for

security i with an expectation of zero and variance ζ2

ξi

The market model is a statistical model which follows the return of any given security to the

return of the market portfolio. It assumes a linear expression based on the joint normality of

the assets returns which represent a potential improvement over the constant-mean-return

model. The method eliminates the portion of the return that is related to variation in the

market model and the variance of the abnormal return. This can thus increase the chances of

predicting the effect of the events. The merit of this model lies on the market model

regression because the higher the regression, the greater the variance reduction of the

abnormal returns (see section 5.1.6. for the equation of the model).

The factor model stems from the attempt to reduce the variance of the abnormal returns by

explaining more of the variation in the normal returns. The market model is a typical example

of the factor model. A multifactor model can be formed by an addition of any factor such as

the industry indexes. The model calculates the abnormal returns by taking the difference

between the actual return and the portfolio of the firms of similar size. However, as Campbell

et al (1997) noted, the use of the multifactor model in event study is naïve. This is because

the marginal explanatory power of the additional factors beyond the market factor is small;

hence there is little reduction in the variance of the abnormal return.

130

The market adjusted model is a restricted market model which has the pre-specified

coefficients. This implies that estimation periods are not required to obtain parameter

estimates. The model is best used to study the under pricing of initial public offer. The model

is most applicable when there is a restriction imposed by limited data availability.

Following Chong et al (2006), this study adopts the market model in measuring the abnormal

returns. A univariate comparison of the cross border and domestic acquirers is undertaken in

order to give an insight on the characteristics and performances of each bank before the

merger. Then, the model is employed in determining the cumulative abnormal returns which

are then used accordingly with other variables in regression, depending on the hypothesis

tested.

5.3 Sources of Data & Sample Period

The starting point in the data sourcing and collection was by comparing the available M&A

databases. The initial lists of M&A deals in the EU were drawn from Thomson Financial

Banker One, Reuters, Bankers Almanac, and Zephyr.

While each of the databases is unique in the type and range of data provided, the nature of the

study required combination of two or more data sources31

at least for the comparison and

because of the nature of the variables. The interest of this study is aligned to bank mergers

returns and dividend policy as such; relevant data relating to shareholders‟ value,

profitability, and dividend policy of the cross border and domestic acquirers were considered

at the same time.

31

Some studies implore different databases to fill in the gap that might be created by omission. Pillof (1999)

used data from IDD M&A Database, Wall Street Journal index, M&A Sourcebook and the Moody‟s Banks and

Finance Manual in constructing detailed merger histories of the banks. Gulger (2003) combined data from two

sources; the CDAX and the Reuters to explore the dividend payout policy among German banks while

Valkanov and Kleimeier (2006) used data from five sources: Zehpyr, DataStream, Edgar, Bankscope and annual

reports in studying cross border M&A. The danger though is that since the databases are provided by different

sources, their approximations and often basis of calculation may create huge disparity in the data. Caution was

therefore exercised in order to ensure that the difference in the databases did not adversely affect the dataset.

131

The relevant M&A list of all commercial banks in the EU was obtained from the Reuters

database from 1997 to 2007. The list comprises basically of commercial banks to commercial

banks deals of domestics and cross border deals. The total number of M&A deals is 62, of

which 42 are cross border and 20 are domestic deals within the EU member countries. Table

5.1 provides a summary of the list.

Since the study is focused on mega M&A deals, coupled with the limited availability of

sample of deals involving only commercial to commercial banks, the criterion on the

minimum value of the included deals was pegged at €500,000 million. The period of the

deals covers from 1997 to 2007. Most of the sample deals were obtained from the Reuters

database while other sources were used to check for consistency (see appendix for list of

domestic and cross border acquirers). The database stores list of the EU commercial to

commercial banks M&A deals, as well as others. In addition, commercial banks32

are easily

identified from other banking institutions such as mortgage and merchant banks.

Some of the financial data of the cross border and domestic acquirer banks were also

obtained from the bankscope database which contains standard and audited balance sheets,

cash flow statements and profit and loss data. These data were used in conjunction with those

obtained in the Reuters and FAME (Financial Analysis Made Easy) databases. In addition,

the daily return before and the after the merger announcement was obtained from the Yahoo

finance.

In accordance with similar studies on mergers and acquisitions, this empirical study applies

some profitability and accounting variables33

such as net interest margin, return on equity,

32

In reality, the difference between commercial banks and other banking institutions seems blurred and obscure

in some databases but this is not so in the Reuters financial database.

33 Different studies on M&A as well as on dividend policy have applied range of variables most of which seek

to measure profitability, efficiency and to shareholders wealth. Pillof (1999) used personnel cost, premises &

fixed assets expenses, total noninterest expenses, capital to assets, loans to assets and deposits to assets in

measuring the changes in performance of shareholders wealth following bank mergers. DeAngelo et al (2006),

Olson & Pagano (2005) and Gugler (2003) used cash dividend, payout ratio (DPS/EPS), dividend yield

(DPS/Mkt value/share), stock dividend,cashholding/TA;TE/TA,sales growth rate, assets growth rate, profit

/sales ratio; dividend/earning ratio; dividend/cashflow,dividend/sales and number of employees in investigating

dividend policy in the banks.

132

cost efficiency ratio, net loans to total assets expenses, equity to total assets, loans to assets,

deposits to assets and total assets, as well as dummies to examine the pre and post merger

performance of the banks. On the other hand, the dividend policy hypothesis is examined

using two dependent variables (payout ratio and dividend yield)34

and other independent

variables such as insider shareholding, stock beta, liquidity, taxes, capital & financial

structures, debt, equity ratio, EPS, ROE, PE , CARs and dummy variables.

5.4 Data and Model

The cross sectional regression model was used to test the profitability and dividend policy

hypotheses. This follows the methods applied by previous studies. For instance, Olson &

Pagano (2006) used the cross sectional model to regress the BHAR (Buy-Hold Abnormal

Returns) over charges in equity, risk, growth and dividend variables in following an M&A;

Delong & Deyoung (2007) applied same model in measuring the post-merger performance

relative to the bank CAR on announcement of merger; Cybo-Ottone & Murgia (2000) used

the cross sectional OLS regression of weighted CARs on set of target variables. Chong, Liu

and Tan (2006) used the cross sectional regression model to test the impact of CARs on some

profitability variables of forced mergers while Goddard et al (2004) compared the results of

the cross sectional model with other models in measuring the EU bank profitability.

The data relating to M&A were built into 4 distinct categories which comprise pre merger

and post merger data for the domestic acquirers and same for the foreign or cross border

acquirers. The data set includes the financial ratios for both domestic and cross border banks

which were extracted from the Reuters and Bankscope databases for the relevant years.

Banks size, growth and investments ratios, debt and financial structures, risk, ownership,

taxes, CTSAR and other such financial variables are used in the comparison of domestic and

cross border acquirers‟ profitability.

Different sets of data are constructed for the purpose of testing the 3 hypotheses. The dataset

for the dividend policy consist of data extracted from their financial statements and other

financial databases and comprises of the 2 dependent variables: dividend payout and dividend

34

Chapters two and three contain detailed discussions on the two dependent variables (the dividend yield and

dividend payout) used as proxy for dividend policy in the study.

133

yield. It also contains the CTSAR used as an independent variable in addition to other

variables for both the cross border and domestic banks.

The results of the regressions are also presented in different models. In most cases, up to 6

models are constructed, each containing different independent variable(s) while controlling

for other factors. The variables are introduced in the models in sequence (sequential

regression) to determine the contributions each makes to the regression model and also

controlling for any exogenous variables. For example, in the testing the profitability

hypothesis, 4 models were constructed each containing the predictor variables. The

coefficients, ANOVA and the result summary are all presented in the models. The dividend

policy hypothesis is presented in 6 models, with each model painting a statistical picture of

the regression results.

5.5 Data Analysis Techniques

The research hypotheses and questions are analyzed using different statistics and econometric

models. The wealth effect and geographical diversification hypothesis was analyzed using the

total cumulative standardized abnormal returns (CTSAR). The process of obtaining the

abnormal returns has been explained in section 5.1.6 and requires some extensive application

of econometric models such as the market model. Both the Z statistics and the p value are

used in evaluating the significance of the CTSAR obtained. However, the CTSAR pooled

data was aggregated over time and bank samples to test the profitability and dividend policy

hypotheses.

The analyses of the profitability hypothesis involved the use of descriptive and parametric

statistics to evaluate the impact of specified independent variables on the financial

performance of the cross border and domestic banks acquirers. The return on equity (ROE)

which is used as a proxy for measuring profitability and is designed to reflect the returns on

the shareholders investments. The independent variables include the net interest income

(NII), equity to total assets (EQTASS), cost to-income ratio (CIR), net loans to total assets

(LOANTAS), Leverage, loan loss provision to net interest revenue (RISK) and the total

assets. The methodology applied follows similar studies such as Rhoades (1998), Vander

134

Vennet (1996), Pillof (1996), Diaz et al (2004), Olson & Pagano (2005) and Campa &

Hernando (2006) in examining the balance sheet ratios.

The NII measures the changes in the ratio of the net interest income to the book assets while

the net loan to total assets gives an indication of the proportion of the balance sheet consigned

to the lending. The loan loss provisions to total loans and the loans provisions to net interest

revenue are indicatives of how prepared the banks are in event of credit losses (Delong &

Deyoung, 2007). Both variables measure the liquidity of the bank. The regression model

follows similar studies in M&A. for instance, Cornett and Tehranian (1992) used ROE and

capital to assets. Altunbas and Marques (2008) included loan to total assets (liquidity proxy),

total assets, while Vander-Vennet (1996) adds loan loss provision to net interest revenue

(credit risk proxy) among the profitability variables.

The profitability hypothesis was tested on unbalanced panel data using the specification

below to evaluate the profitability model equation (See section 6.4 for the explanation of the

equation variables):

Profitabilityi = + β1 NIIi + β2SIZEi +β3EQTASSi +β4CIRi + β5LOANTASi +β6RISKi

+ β7PROVTASi + β8CTSARi + β9Dummyi + ei

Where: NII is the net interest income, SIZE is the total assets, EQTASS is the equity to

total assets ratio, CIR is the cost to income ratio, LOANTAS is the loan to total assets,

which is a proxy for liquidity, RISK is the loss provision to net interest revenue, indicating

the ratio of risky lending, PROVTAS is the Loan provisions to total assets; this represents the

liquidity of the bank, CTSAR is the cumulative total standardised abnormal return, proxy

for M&A, Dummy 0 for pre-merger and 1 for post-merger events + e is the error term.

The results are then presented in tables and analyzed using their ANOVA values. The

standardized beta coefficients were used to estimate the contributions of each of the variables.

The variables with higher standardized beta coefficients are adjoined to exert more influence

and thus affect the models. The p values of the variables are set at 5% and 1% degrees of

freedom to test their significance.

135

The sequence is repeated for both the domestic and cross border acquirers. The difference in

the significance of results of both samples (cross border and domestic acquirers) and

highlights the common differences in their respective performances following the M&A

events. A regression combining both sets of data is obtained using the appropriate dummy

variables distinguishing between domestic and cross-border M&A, as well as cash and non-

cash acquisitions.

The univariate statistics and hierarchical regression were used to test the CTSAR and

dividend policy hypothesis which follows similar pattern as with the profitability analyses.

Two dependent variables; the dividend yield and payout are used as proxies for dividend

policy and regressed against the CTSAR and other independent variables of the cross border

and domestic acquirers. The predictor (independent) variables include the beta, liquidity,

insider shareholdings, taxes, equity to total assets, total assets, debt to equity, earnings per

share, return on equity, price earnings and the dummies variables35

.

A further check on the results identified in the regression is undertaken to convincingly

establish that the effects exhibited by the predictor variables on the dividend policy of the

merged banks were not just by chance. Thus, the Granger Causality test was applied to

rigorously examine that there is a temporal cause-and –effect relationship among the CTSAR

and the dividend policy variables. The equation of the CTSAR and dividend policy

hypothesis is as stated36

:

DivPolicy = β0 – β1Betai + β2Liquidityi - β3Insideri + β4Taxi - β5Cap&FnSti + β6Sizei +

β7Profiti – β8Growthi + β9CARsi + ξi

where: Beta is the estimated beta coefficient of the banks (with negative sign to indicate its

expected effect), Liquidity is the availability of physical cash in the bank measured as the

35

Due to collinearity problem, only 2 of the dummy variables, the DumGeoCARs, which measures the domestic

acquirer‟s CTSAR and the PreDumTass, which measures the premerger total assets of the domestic acquirers

were finally used in the analyses. The DumGeoCARs takes the value of 1 for cross border acquirers and 0 for

the domestic acquirers. The issue of collinearity has been explained in section 5.7 (test of multicollinearity).

36 See section 7.2 for the explanation of the variables used in the dividend policy regression model. The

mathematical sign(s) before the variable indicates the expected outcome of the result of the variable.

136

dividend/net cash operating, Insider is the percentage of insider shareholdings in the

acquiring banks (with expected effect being negative) , Tax is the total tax liabilities of the

banks as well as the relevant tax ratios, Cap&FnSt the bank‟s capital and finance structures

measured by the Debt/Equity ratio (capital structure) while the finance structure is debt +

equity/total assets, Size the natural log of the total assets is used as the proxy for size of the

bank, Profit is the profitability of the bank as measured by the ROE and EPS, Growth is the

price earnings (PE),(with negative expected effect on dividend policy),CTSAR is the

Cumulative total standardized abnormal returns (CTSAR) is a proxy for M&A and ξ is the

error term.

The inclusion of these variables in the dividend model follows the pattern of other studies.

For instance, Gugler (2003) and Casey & Dickens (2000) models include percentage of

institutional ownership (insider) as well as bank beta and tax variables. Pattenden and Twite

(2008) used profitability, size, taxes, and credit risk in their dividend policy models. Other

studies that have applied similar variables include McManus et al (2004) and Anil and

Kapoor (2008) respectively.

5.7 Test of multicollinearity

The regression variables were subjected to a multicollinearity test in order to ensure that the

results do not suffer from the mishaps of multicollinearity. The problem usually occurs

among explanatory variables that have linear relationship. Multicollinearity represents the

extent to which the effect of any variable can be explained by the other variables in the

analysis. It diminishes the ability to determine the right effect of the variable on the model

(Hair et al, 2005). In practice, the symptoms of multicollinearity include false signs of

coefficients, smaller than expected t-values and larger than expected insignificant p values

(Albright, Winston & Zappe, 1999).

The preliminary analyses include the collinearity diagnostics test; the residual statistics and

The Cook‟s Distance in the Residual table were cross checked to ensure that the maximum

value is not more than one (1). Tabachnick & Fidell (2001) suggest that values more than one

(1) may constitute some problems in the analysis.

137

Since the coefficient represents the impact of the independent variables on the model in

addition to the effects of the other variables in the regression equation, it was imperative to

ensure that there is an established correlation among them. Tabachnick & Fidell (2001)

advise that a correlation coefficient above 0.7 or below 0.3 may be worrisome. Therefore,

variables that show signs of „improper correlations‟ were eliminated from the list.

Some data were deleted from the panel either because they are outliners or were particularly

spurious37

.

The scatter plot was used in detecting the outliers while the residual and collinearity statistics

were employed to examine any further multicollinearity in the result. The Coefficient Tables

6.6 and 6.15 (Chapter 6) contain the collinearity statistics (see the top right section of the

tables) to help check that the results are free of any multicollinearity. Multicollinearity

problems are usually detected by the VIF (variance inflation factor) and the Tolerance values.

The Tolerance value measures the extent of the variability of the specified independent

variables which are not measured by the rest of the independent variables and calculated as

1 – R2 . Any value less than 0.10 indicates high correlation with other variables and

indicating a multicollinearity issue. The VIF is the inverse of the Tolerance value. Pallant

(2006) advised that the VIF value should be below 10.00 for a reliable result. All the

variables included in the analyses meet up with the criteria. .

37

The data were collected and compiled from different sources. The major sources include the websites of the

banks, Bankscope, Reuters, Thompson Financial and financial analysis made easy (FAME). In some cases, the

differences in the values of data are so wide that the data is treated with suspicion or totally eliminated as an

outlier in the analyses. Some data were removed due to the multicollinearity problem. One of the dependent

variables, dividend payout was eliminated from the panel due to the issue of collinearity with the predictor

variables.

138

Chapter Six

Data Analyses and Discussion of Results

6.0 Introduction

This chapter presents the empirical analysis undertaken in the study and discusses the results

obtained. Many studies on mergers and acquisitions have over the decades preoccupied the

financial literature, but the approaches to the studies have often been methodological and

focused on identifying gains to shareholder wealth. This study has adopted the approach of

pairing the subject of M&A not only to shareholder wealth and the profitability of acquirers

but also to the controversial issue of dividend policy.

Different studies have diagnosed a multitude of reasons for mergers and acquisitions.

Financial institutions, particularly banks, are very central to the economy of countries and

therefore decision to undertake a merger is usually considered with utmost seriousness.

Chapter 4 discusses some reasons for M&A and problems and challenges of merged entities.

For the empirical work, this study has focused on issues related to domestic mergers as well

as foreign or cross border mergers. To this effect, three testable hypotheses, discussed in

Chapter 1, are tested and analysed in this and the next chapter. The hypotheses and the

expected outcomes are listed in Table 6.1 below.

Table 6.1 Research Hypotheses, Proxies and their Expected Outcomes.

Hypothesis Proxies Expected outcome

The wealth effect & CTSAR: Cumulative total standardised Positive (cross border)

geographical abnormal returns of both domestic & Negative(domestic)

diversifications cross border merged banks

hypothesis

The profitability hypothesis of cross

border and domestic acquirers. ROE: Return on equity Positive (cross border)

Negative (domestic)

Dividend policy hypothesis - Dividend Yield & Positive (both)

Dividend Payout

139

6.1 The Wealth Effect and Geographical diversification Hypothesis

This section tests the wealth effect and geographical diversification hypothesis using the

standardized abnormal returns of the cross border and domestic acquirers. It also discusses

the result obtained from the market model of the event study. The event study method was

described in Chapter 5 of the research methodology.

The AR was standardised to cater for the different degree of event impact. This is done by

weighting each abnormal return by their standard deviation. The purpose of the

standardization is to ensure that each abnormal return has the same variance (Serra, 2002).

Thus, by dividing each firm‟s abnormal residual by the standard deviation over the estimation

period, each residual will have an estimated variance of 1 and thus defined by the equation:

= ---------------------------------------------------- 1

Where = SAR for firm j at time t.

= AR for firm j at time t.

=

Square root of the variance of the AR for firm j at time t = the standard deviation of the AR

for the firm j at time t.

6.1.1. : The wealth effect and geographical diversification hypothesis: In contrast to

domestic acquisition, cross border acquisitions will not create more value. Where

diversification and possible reduction in financial distress is achievable, foreign acquisitions

are not expected to outperform domestic acquirers.

The above hypothesis was tested using the results of the cumulative total standardised abnormal

returns (CTSAR) of both the cross border and domestic bank M&A, as shown in Table 6.2.

The shareholders value, proxied by CTSAR of the cross border (international) acquirers is compared

with the domestic acquirers.

140

The results obtained provide empirical insight on the pattern of the returns. The cross border

acquirers show significant negative CTSAR, while the domestic acquirers have insignificant

CTSAR. Using 61 days window (30 days before and after plus the event day), it can be

established that many of the significant returns of the cross border occurred after the

announcement of the M&A. Only few significant returns occurred before the mergers

announcement date. The domestic banks show no significant returns throughout the event

window.

Out of the 21 significant results, 5 occurred prior to the mergers while 16 occurred after the

merger event. Some reasons can be adduced as to why there are more significant abnormal

returns of the cross border banks after the announcement date. The public confidence may

have been affected which may trigger upward shock in the share price of the banks. Since the

cross border bank samples are made up of large international banks, it is apparent that their

targets would be similar banks with substantial market share. The news of the M&A would

either build or decrease the confidence in the public. Table 6.2 shows the significant results

of the cross border and domestic acquirers.

141

Table 6.2 Result of the Cumulative Total Standardised Abnormal Returns of Cross border &

Domestic M&A

Event

days

Cum

TSAR+

Z-stat P-Value Event

days

Cum TSAR Z-Stat P-Value

-30 -12.25

(-0.74)#

-2.86 (-

0.16)

0.004*

(-0.87)

18 -34.14

(-10.35)

-2.80

(-0.50)

0.005*

(0.61)

-26 20.78

(1.87)

2.17

(0.38)

0.03*

(0.71)

21 -27.37

(-5.86)

-2.18

(-0.32)

0.03*

(0.75)

-25 29.02

(5.14)

2.76

(0.91)

0.006*

(0.36)

22 -33.82

(-7.39)

-2.37

(-0.38)

0.018*

(0.71)

-12 -38.19

(-6.01)

-3.11

(-0.49)

0.002*

(0.63)

23 -25.37

(-8.85)

-2.07

(-0.49)

0.038*

(0.62)

-11 -25.38

(-6.3)

-2.89

(0.33)

0.004*

(0.74)

25 -28.09

(-9.101)

-2.43

(-0.38)

0.015*

(0.70)

10 -28.03

(-4.63)

-2.27

(-0.33)

0.023*

(0.74)

26 -26.61

(-11.63)

-2.68

(-0.52)

0.008*

(0.60)

11 -32.03

(-5.57)

-2.23

(-0.3)

0.026*

(0.77)

27 -30.57

(-14.62)

-2.47

(-0.58)

0.014*

(0.56)

12 -34.81

(-8.5)

-2.66

(-0.45)

0.008*

(0.65)

28 -39.90

(-16.86)

-3.03

(-0.69)

0.002*

(0.49)

13 -38.6

(-12.9)

-2.83

(-0.68)

0.004*

(0.5)

29 -38.41

(-18.33)

-2.96

(-0.75)

0.003*

(0.45)

14 -27.3

(11.43)

-2.51

(-0.59)

0.012*

(0.55)

30 -25.23

(-21.44)

-2.25

(-0.98)

0.024*

(0.33)

15 -28.2

(11.93)

-2.79

(-0.66)

0.005*

(0.51)

16 -24.7

(11.61)

-2.88

(-0.66)

0.003*

(0.51)

17 -41.5

(-8.6)

-3.42

(-0.47)

0.006*

(0.64)

*Significant at 0.05. Only significant returns are shown in the table.

#Figures in parentheses are for domestic M&A

+CTSAR (cumulative total standardized abnormal returns).

The graphs of the abnormal returns are represented in Figures 6.1 and 6.2. Note the rise and fall

in the slope of the curves of the cumulative total standard abnormal returns after the event date.

Figure 6.1 illustrates the curve of the cross border acquirers. The graph slopes down deep into

negative CTSAR to about -41.5%, just after the mergers but with an initial rise after the 10th

day

before falling back. But prior to the M&A, the CTSAR was 29.02% which occurred on the 25th

day before the merger news infiltrated the market. The abnormal returns were at peak just before

the announcement date when the information filtered into the market. But fell slowly before day

zero.

142

Figure 6.1 Graph of the Cross border Cumulative Total Standardized Abnormal Returns

The graph depicts the nature of the CTSAR of the cross border acquirers. There are more

significant results as shown in the Table 6.2, most of them are negative abnormal returns. The

news of the mergers triggered an initial increase but with a subsequent fall in the cumulative

total standardized abnormal returns before the event day. The CTSAR of the cross border

continues to fall even after the mergers.

Figure 6.2 gives a similar illustration of the shape of the graph of the domestic mergers.

However, the graph of the domestic CTSAR shows a more sustained positive growth. This is

paradoxical as the growth shown in the graphs is not significant. Note also that the news of the

merger triggered an initial growth in the curve of the CTSAR before its lowest fall on the event

day to up to 20%. The fall was reversed by a steady growth after the event day before eventually

steeping downwards.

143

Figure 6.2 The Graph of the Domestic Cumulative Total Standardized Abnormal Returns

The graphs do not reflect the significance of the results but shows the growth trend in the

abnormal returns of the bank domestic acquirers. It shows that the slopes of the domestic

acquirers move faster above zero (positive) than the cross border acquirers. Though, the latter

has more negative significant results than the former.

Domestic acquirers have been adjudged to possess more knowledge of the local market than

foreign acquirers. This may explain the jump in the CTSAR before the mergers

announcement news infiltrated the market prior to the eventual fall. Das & Sengupta (2001)

averred to asymmetric information as an underlining variable for the disparity in the growth

of the abnormal returns of the domestic and foreign acquirers. The domestic banks are likely

to know more about the preferences of domestic consumers than would the cross border

acquirers. Their presence in the domestic market in related lines of activity often results in

greater experience regarding household income and expenditure patterns, and household

preferences for new products and brands. All of this adds up to private information in the

possession of the domestic banks and their advantageous position in the market during the

144

merger period. Some studies have asserted that the growth in abnormal returns is hugely

influenced by the method of payment and the influence of tax in the M&A deal38

.

The result obtained from the event study in relationship to the wealth effect and geographical

diversifications hypothesis confirms that although the cross border acquirers have more

negative significant CTSAR, on aggregate basis, the domestic acquirers bring relatively better

returns to their shareholders. Some of the returns of the cross border acquirers are in high

negative while those of the domestic are in relatively low negative CTSAR.

Table 6.3 Cumulative Total Abnormal Returns (CTSAR) in Quartiles

The table shows the CTSAR in quartiles of the cross border and domestic acquirers. The 1st and 2

nd quartiles are in

the pre merger periods while the 3rd

and 4th

quartiles are in the post merger periods. The quartiles are split into 60days

of the event window.

1st Quartile

(-30+0 ) days

2nd

Quartile

(-15+0) days

3rd

Quartile

(0+15) days

4th

Quartile

(30+0) days

Pre-cross

Pre

Domestic Pre-cross

Pre

Domestic Post-cross

Pre

Domestic Post-cross

Pre

Domestic

-12.26 0.07 -21.84 -0.69 -38.70 -13.55 -24.66 -11.61

0.41 4.30 -22.46 0.44 -34.43 -4.07 -41.48 -8.60

13.23 4.71 -32.38 -4.20 -37.79 -6.79 -34.14 -10.35

-9.28 2.07 -38.19 -6.01 -39.14 -6.53 -25.97 -6.62

20.78 1.87 -25.38 -6.31 -41.11 -0.88 -34.34 -7.57

29.02 5.14 -28.92 -7.36 -32.65 0.80 -27.37 -5.86

20.32 7.08 -13.86 -7.96 -35.47 4.41 -33.82 -7.39

13.85 8.98 -24.81 -6.95 -36.70 -0.65 -25.37 -8.85

19.45 3.87 -5.44 -11.79 -39.20 -4.98 -21.50 -8.99

20.18 2.54 -11.74 -11.67 -32.30 -4.63 -28.09 -9.10

17.00 3.03 -15.14 -17.83 -32.03 -5.57 -26.61 -11.63

22.80 5.47 -27.68 -15.16 -34.81 -8.50 -30.57 -14.62

9.10 9.16 -29.84 -17.78 -38.60 -12.90 -39.90 -16.86

15.61 2.87 -39.77 -16.42 -27.32 -11.43 -38.41 -18.34

-4.35 2.24 -41.58 -20.09 -28.18 -11.93 -25.23 -21.44 AVG

CTSAR 175.43 63.40 -379.03 -149.75 -528.43 -87.21 -457.47 -167.83

38 Huang and Walkling (1987) observe that the larger abnormal returns for cash acquisitions are consistent with

tax explanation. That is, target shareholders are taxed for capital gains in cash mergers, but the taxes are

deferred in mergers involving stock transactions. A cash offer would have to be larger than a stock offer to

offset the tax liability. However, Davidson & Cheng (1997) conclude that a cash offer is not inherently more

valuable than a share exchange that offers the equivalent amount of shares. Cash contains no additional

information, provide no additional value, and do not seem to alleviate the asymmetric information problem any

more than offers made with payments of stock. Cash offers are the source of larger returns only because cash

targets received larger payments from bidders and concluded that the link between abnormal returns and

payment method is indirect, rather than direct.

145

Table 6.3 shows the CTSAR values in quartiles. The pre M&A periods constitute the first

two quartiles while the post M&A periods are in the last two quartiles. The result indicates

that the 1st quartile produces the highest CTSAR of 174.43 for cross border acquirers‟

shareholders followed by the 2nd

quartile with a CTSAR value of 63.40 for domestic

acquirers.

The worst period for the cross border acquirers is in the 3rd quartile with a CTSAR of

-528.43. The domestic acquirers recorded their lowest CTSAR in the last quartile with a

CTSAR value of -167.83. The results therefore corroborate the earlier argument that the

CTSAR of the cross border acquirers fall at a greater speed than those of the domestic

acquirers. This thus reduces the wealth effect or benefits of their shareholders from such

M&A.

Thus, based on the result of the acquirers in Table 6.2, the null hypothesis that the cross

border acquirers do not create more values than the domestic acquirers is thereby accepted.

The result obtained is though similar to some other studies on cross border and domestic

acquirers but it uses the CSTAR as against the traditional cumulative abnormal return which

is common in many studies. The findings concurs with Lensink & Maslennikova (2008) that

domestic transactions create more value than the cross border. Using a window of 41 days,

they found that the cross border acquirers never created value to their shareholders. Mangold

et al (2008) concludes that cross border acquirers destroy wealth while domestic acquirers

create value. Some studies (Anand, Capron & Will, 2005 and Conn et al, 2005) have also

found that both acquirers exhibit similar pattern or have no statistical difference in their

abnormal returns to the shareholders.

146

6.2 Descriptive Analysis of the Financial Performance of the Cross Border and

Domestic Bank Acquirers

The descriptive analysis of the financial performance of both domestic and cross border

acquirers is measured by their profitability. The summary statistics is presented in Table 6.3

and makes comparison of the cross border and domestic acquirers using key profitability

variables as well as their CTSAR.

Table 6.4 Summary Statistics of the Abnormal Returns and Merger Performance of the Domestic

and Cross Border Banks for the period 1997 -2007.

The table presents the summary statistics of the acquirers‟ financial performance using key ratios. The profitability is measured

using the traditional measures of Return on Equity (ROE) which is a key profitability variable. Others include the Net interest

income (NII) and the Cost to- income ratio or the cost to income ratio (CIR) which show the profitability arising from the savings

due to the banks efficient operations. The Equity/total assets (EQTASS) and the) measure the contribution of the capital structure

of the bank to its performance. It also explains the profits arising from the quality of the banks‟ assets. Net loans to total assets

(LOANTAS) and the Loan loss provision to total assets (PROVTAS) measure the liquidity of the banks. They capture the banks

financial position, investments strategy and funding methods. The Loan loss provisions to net interest revenue (RISK) pictures the

risky nature of the banks loan and the net effect on its liquidity and profitability while the total assets indicates the size of the

banks.

Panel A: Pre-merger performance of domestic banks

N Minimum Maximum Mean Standard Deviation

Return on Equity (ROE) 86 -18.36 34.06 9.35 8.28 Net interest Income (NII) 85 -10.01 4.85 1.73 2.20 Equity/total assets(EQTASS) 85 1.71 15.14 6.74 3.95 Cost to-income ratio(CIR) 86 41.44 94.72 66 18.48 Net loans to total assets(LOANTAS) 81 5.41 85.21 52.18 19.80 Loan loss provision to total assets (PROVTAS )

83 0.80 8.78 2.40 1.23

Loan loss provisions to net interest revenue (RISK) 88 -1.51 99.42 19.57 15.27

Total assets 83 1,129.80 1,681,490

218,426.52

390,800.02

CTSAR 30 -20.09 9.16 -2.88 8.68

Panel B: Post-merger performance of domestic banks N Minimum Maximum Mean

Standard Deviation

Return on Equity (ROE) 130 -42.80 23.76 8.92 10.79

Net interest Income (NII) 129 0.68 4.51 2.06 0.94

Equity/total assets(EQTASS) 128 1.42 11.13 6.15 2.54

Cost to -income ratio(CIR) 130 30.85 116.83 62.02 15.65

Net loans to total assets(LOANTAS) 129 10.65 94.49 60.28 20.32

Loan loss provision to total assets (PROVTAS) 129 0.71 15.30 4.08 3.44

Loan loss provisions to net interest revenue (RISK) 128 -14.31 26.11 25.45 32.34

Total assets 132 2,587.40 2,766,076.00 526,142.

09 758349.62

CTSAR 30 -21.44 4.41 -8.50 5.60

147

The table makes a comparison between the financial performance of the domestic and cross

border acquirer samples in the study. Using key performance variables, the result shows a

reduction in the ROE, EQTASS and CIR of the domestic acquirers. The mean of the ROE fell

after the M&A from 9.35% to 8.92% and the EQTASS reduced from 6.74% to 6.15% while

the CIR fell from 66% to 62.02% after the merger process.

The decrease in ROE indicates that on average, the market investors did not efficiently price

the bank mergers during the period. This is a phenomenon based on previous merger

performance studies. This decline is not surprising as on average, the post merger financial

performance of banks as measured by the ROE rarely improves (Delong & DeYoung, 2007).

The fall of EQTASS is very marginal and therefore does not paint a gloomy picture in terms

of the equity structure of the merged entity.

Panel C: Pre-merger performance of Cross border banks N Minimum Maximum Mean

Standard Deviation

Return on Equity (ROE) 129 0.09 45.92 14.71 7.30 Net interest Income (NII) 130 0.72 3.71 2.27 0.69 Equity/total assets(EQTASS) 132 2.59 11.54 5.52 1.75 Cost to –income ratio(CIR) 127 45.88 93.52 64.40 9.20 Net loans to total assets(LOANTAS) 130 26.76 82.18 52.27 11.85 Loan loss provision to total assets (PROVTAS )

130 0.28 5.97 2.68 1.34 Loan loss provisions to net interest revenue (RISK)

126 -4.46 52.37 15.71 9.86 Total assets 131 85,966 288,454 153,414.44 83,444.89

CTSAR 30 -41.58 29.03 -6.79 21.95

Panel D: Post-merger performance of Cross border banks

Return on Equity (ROE) 326 -18.47 28.72 11.20 9.36

Net interest Income (NII)) 331 0.49 4.33 1.54 0.79

Equity/total assets(EQTASS) 302 2.57 10.67 5.38 2.25

Cost to –income ratio(CIR) 315 44.56 112.67 64.83 12.04

Net loans to total assets(LOANTAS) 325 27.33 87.02 51.86 15.42

Loan loss provision to total assets (PROVTAS) 326 0.07 4.87 2.39 1.43

Loan loss provisions to net interest revenue (RISK) 333 -1.78 54.01 16.97 12.90

Total assets 321 6,427.80 968,733 393,055.04 284,477.90

CTSAR 30 -41.48 -21.50 -32.86 5.70

148

CIR is the cost to-income ratio and shows a substantial decline. This is an indication of more

efficiency in the control of costs and expenses. Since mergers are often set out to achieve

palpable reduction in costs and increase in efficiency, thus reduced expenses would be

expected to translate into higher profitability though reported incomes do not often reflect the

actual earnings (Pillof, 1996). The table shows a significant increase in the NII of domestic

acquirers from 1.73% to 2.06%. The NII measures the difference between the revenues on the

assets and the cost of servicing the liabilities. This increase could be attributed to a higher

interest charge on loans which increased substantially after the M&A. It represents the interest

payments received by the banks on loans and those made by the banks on customers‟ deposit

and reflects the success of core intermediation activity (Vander Vennet, 1989).

The variables LOANTAS, PROVTAS, RISK and the total assets of the domestic acquirers

also shows increases after the merger process. LOANTAS rose from 85.21% to 94.49%,

PROVTAS 8.78% to 15.30% while the RISK increased from 19.57% to 26.11% respectively.

LOANTAS shows an increase in the amount of loan held by the banks which can be

explained by the proportional increase in NII and depicts the assets quality of the banks

(Pillof, 1996).

The variables PROVTAS and RISK measure the exposure to credit risk of the banks after

mergers. Banks use loan loss provisions to create reserves to cushion the expected losses in

their loan portfolios (Perez et al, 2006). The increases are indications that the mergers invoked

anticipations in loan losses to customers and thus creating further provisions in bad and

doubtful debts after the mergers. However, these increases are by no means accidental but

suggest that the mergers have a significant shift in the debts and investment strategy of the

banks.

The post merger profitability variables of the cross border acquirers also show similar pattern

except in the variables NII, CIR, LOANTAS and PROVTAS. NII fell from 2.27% to 1.54%,

LOANTAS dropped from 52.27% to 51.86% while PROVTAS decreased from 4.08% to

2.39% respectively. However, CIR records a marginal increase from 64.40% to 64.83%. The

fall in NII has an adverse effect on the overall profitability of the bank. The net interest

income is presumed to be inversely related to the level of interest rates. Thus the decrease can

be attributed to as the variation in net interest income resulting from interest rates volatility

(Albert & Pamela, 2007).

149

CIR of the domestic acquirers shows a marginal increase; which represent less managerial

efficiency in the control of costs. It measures how effective the banks cost operations are. It

remains the traditional measure of banks efficiency and measures non-interest expenses as a

proportion of operating revenue. The costs include salaries, technology, buildings, supplies,

and administrative expenses. While revenue includes net interest income (interest revenue less

interest expenses) plus fee income revenue (Finerran, 2006). Thus the increase in CIR is a

negative signal in the management of cost.

Studies (Ahmed at el, 1989; Beaver et al, 1989) have shown that the decrease in LOANTAS

and PROVTAS has a signal effect and portrays that the banks as weak. This is consistent with

the signalling hypothesis and provides support that LOANTAS and PROVTAS are used as

capital management and smoothing techniques. In comparison with the domestic acquirers,

cross-border acquirers show weak financial position in their LOANTAS and PROVTAS. The

increase in the total assets of both the domestic and cross border acquirers is not surprising as

mergers expand the size of the banks assets base.

The mean of the CTSAR of the domestic acquirers sloped down from -2.88 to -8.50 after the

M&A. This is better when compared with those of the cross border which plummeted deep

into negative from -6.79 to -32.86. This result leans support to the rejection of the wealth

effect and geographical diversification hypothesis (H1,) and provides further evidence that

domestic acquirers at aggregate create more value to their shareholders than their cross

border counterparts.

6.3 Performance Analysis of the Domestic Acquirers

The starting point of the performance analysis was to ascertain the extent of correlation

among the financial performance variables (ROE, NII, CIR, EQTASS, LOANTAS,

PROVTAS and total assets used as a proxy for SIZE). A correlation matrix was performed

using the Pearson product moment correlation coefficient. Tables 6.5 and 6.6 shows the

results of the correlation matrix of the pre and post M&A performance variables of the

domestic acquirers used in the study.

150

The pre merger profitability correlation matrix table also show some significant correlation

among the profitability variables and the ROE. Both the NII and the EQTASS variables have

a correlation of -0.223 respectively with the ROE. The LOANTAS and RISK correlate with

the ROE at -0.193 and 0.196. The negative correlations of the variables imply that the higher

the profitability of the domestic acquirers, the less is the NII, EQTASS and LOANTAS

values. RISK has a positive relationship with ROE; implying that a high performance attracts

a higher RISK. This findings leans support to previous profitability studies (Zhou & Wong,

2008; Delong & Deyoung, 2007, Deyoung, 1999).

However, the post merger profitability correlation matrix Table 6.6 shows that the CTSAR

does not have significant correlation with the profitability variables of the domestic acquirers

after the M&A. The implication of the result is that while the merger announcement could

influence the financial performance of the domestic acquirers, it does not have similar

influence after the merger. Interestingly, there are remarkable changes in the correlations of

Table 6.5 Correlation Matrix of Cumulative Total Standardised Abnormal Returns (CTSAR) with the Pre Domestic

Performance Variables

Variables ROE NII EQTASS CIR LOANTAS PROVTAS RISK

SIZE CTSAR

ROE 1.000

NII -.035 1.000

EQTASS -.212 .762** 1.000

CIR -.472** .207 .264 1.000

LOANTAS -.012 .118 .220 -.450** 1.000

PROVTAS -.561** .304

* .492

** .258 .080 1.000

RISK -.733** .025 .039 .099 .177 .480

** 1.000

CTSAR

-.055 -.027 -.028 -.045 -.023 .069 .062 .053 1.000

**significant at 0.01 levels, *significant at 0.05 levels.

151

NII, EQTASS and LOANTAS with ROE as all diminishes to insignificant values after the

M&A.

RISK shows a significant negative correlation of -0.733 with ROE after the M&A. This

indicates that there is a relationship between profitability and risk, as high risk investments

will commonly be expected to yield a higher profit. CIR and PROVTAS indicate a high

negative correlation with ROE at -0.561 and -0.733 respectively. This is not surprising as the

liquidity of the bank (proxied by PROVTAS) and effective costs savings (CIR) are key

variables in the banks financial performance. This finding confirms existence of a

relationship between bank liquidity and profitability. In fact, Goddard, Molyneux & Wilson

(2004b) assert that banks that maintain high liquidity are more than likely to record a low

profitability. This also impairs growth as cash assets are not invested into productive projects

and therefore rendered dormant.

ROE NII EQTASS CIR LOANTAS PROVTAS RISK SIZE CTSAR

ROE

1.000

NII -.223* 1.000

EQTASS -.223* .507

** 1.000

CIR -.142 -.002 .185 1.000

LOANTAS -.193* .054 .143 -.086 1.000

PROVTAS -.174 .166 .122 .045 .051 1.000

RISK .196* -.011 -.022 -.008 -.015 -.157 1.000

SIZE .001 -.060 -.399** .140 -.407

** .226

* -.074 1.000

CTSAR .396 -.119 -.754** -.334 -.738

** .181 -.157 .239 1.000

**significant at 0.01 levels, *significant at 0.05 levels.

Table 6.6 Correlation Matrix of the Cumulative Totals Standardised Abnormal Returns (CTSAR) with the Post

Domestic Performance Variables

152

In summary, M&A influences the performance of NII, ETQASS, CIR, LOANTAS and

PROVTAS of the domestic acquirers. However, it would appear that the decrease in the

profitability reduces CTSAR of the banks.

6.4 Test of Hypothesis on Financial Performance of Bank Acquirers

H2: The financial performance of cross border acquirers is significantly correlated with

M&A, and significantly higher than the domestic acquirers.

Given their access to capital through their foreign partners and investments, cross border

bank acquirers are expected to outperform domestic acquirers. The cross sectional regression

model is used to test the above hypothesis and examine the contributions of the independent

variables to the financial performance of the merged banks.

Using profitability (proxied by ROE) as the dependent variable, the regression equation is

more generally specified as:

= - β1 NIIi + β2SIZEi +β3EQTASSi +β4CIRi + β5LOANTASi +β6RISKi

+ β7PROVTASi + β8CTSARi + β9Dummyi + ei

Where:

Β0 = intercept term

NII = Non interest Income which measures the percentage of profits from off-balance sheet

operations.

SIZE = proxy for total assets

EQTASS = equity over total assets

CIR = Cost to- income ratio, indicating the efficiency in management of expenses relative to

income.

LOANTAS = Loan deposits over total assets to show the percentage of deposits that are

returned to the banks through loans.

RISK = Loan loss provision to net interest revenue, indicating the ratio of risky lending.

PROVTAS = Loan provisions to total assets; this represents the liquidity of the bank.

CTSAR= Cumulative total standardised abnormal returns, proxy for M&A.

Dummy = 0 for pre-merger and 1 for post-merger events.

ε = error term.

153

6.4.1 Performance Analysis of Domestic Acquirers

The above specifications39

were used to determine the contributions of the SIZE, EQUITY,

LOANTAS, RISK, PROVTAS, CTSAR and dummy variables on the financial performance

of the merged banks. The results are shown in Tables 6.7.

The table provides an indication of the significance of the regression used in assessing the

domestic acquirers‟ financial performance. Model 1 shows that two variables (NII and

SIZE) contribute 15.1% to the performance variance and Model 2 makes a 5.2% contribution

to the variance. Models 3 and 4 explain 33.4% and 49.1% of the variances of the financial

performance of the domestic acquirers respectively. These are indicated by the value of the

Adjusted R Square which are used in determining the level of effect exerted by all the

variables included in that particular model .

39

The essence of the segmentation was to assess the impact of each of the independent variables on

performance. At the same time, the multiple regressions allow for ease of control for some factors or variables

while assessing the rate of impact on the overall result by some other variables of interest (Pallant, 2005).

154

Table 6.7 Model Summary of the Hierarchical Regression of Financial Performance of

Domestic Acquirers The model summary shows the beta value and the t-values (in parentheses) of the profitability variables.

Independent variables for the first model consist of the NII and the SIZE. In the second model, an additional

variable; EQTASS was added. Three further independent variables i.e. CIR, LOANTAS & RISK were also

included in the third model. While the fourth model has in addition to the above, PROVTAS, CTSAR and

DUMMY variables.

Variables Model 1 Model 2 Model 3 Model 4 Tolerance VIF

NII -0.224* -0.164 -0.157 -0.141 0.692 1.445

(-2.318)# (-1.456) (-1.424) (-1.239)

SIZE -0.142 -0.12 -0.137 -0.117 0.854 1.171

(-1.472) (-1.216) (-1.417) (-1.144)

EQTASS -0.118 -0.077 -0.128 0.528 1.892

(-1.026) (-0.682) (-0.984)

CIR -0.378 -0.381** 0.736 1.358

(-1.859) (-1.940)

LOANTAS -0.095 -0.101 0.813 1.230

(-0.989) (-0.673)

RISK 0.174 0.189 0.967 1.035

-1.829 -1.757

PROVTAS -0.403** 0.490 2.040

(-0.764)

CSTAR -0.208 0.821 1.115

(-1.411)

DUMMY 0.006 0.711 1.406

-0.054

Constant 4.478** 4.594** 4.591** 3.537**

Mean Sq 326.08 247.75 242.83 188.92

F-Value 13.77 12.86 92.97 97.28

Sig. ANOVA 0.027* 0.041* 0.000** 0.000**

Change Statistics

R Square 0.17 0.08 0.356 0.496

R Sq Change 0.17 0.082 0.377 0.406

Adj. R Sq. 0.151 0.052 0.334 0.491

Std. Error 9.306 9.303 9.046 9.11

Sig.F Change 0.027* 0.307 0.000** 0.000**

Df1 2 1 3 3

Df2 100 99 86 94

# Figures in parentheses are t-values,*significant at 0.05 level. **significant at 0.01 level.

155

The Model Summary also shows that Model 3 variables make a high contribution of 35.6% to

the financial performance as indicated by the R Square. Three variables: CIR, LOANTAS

and RISK were introduced while controlling for NII, SIZE and EQTASS. The R Square

Change indicates that the 3 variables introduced in the model made an additional input of

37.7% to the performance variance. This is quite significant as the Sig. F Change is p=0.007.

It is thus apparent from the Adjusted R Square that Model 4 exerts the most influence on the

performance of the acquirer banks as it makes the highest contribution of 49.1% of the variance.

Three variables, PROVTAS, CTSAR and DUMMY, are additional variables of interest. The R

Square Change indicates that these variables explain an additional 40.6% of variance in the financial

performance after controlling for NII, SIZE, EQTASS, CIR, LOAN and RISK.

The overall regression result is significant as confirmed by the Sig. F Change value. Table 6.6 of the

ANOVA statistics also confirms the significance Model results.

The beta coefficient values in Table 6.6 provide further insight on the extent of contribution made by

each of the variables of the domestic acquirers. This is determined by the beta40

value of the variables.

The beta values of PROVTAS, CTSAR and DUMMY in Model 4 are 40.3%, 20.8% and 0.6%

respectively. The results clearly indicate that PROVTAS makes the highest significant

contribution to profitability. Thus, PROVTAS is confirmed as the most significant variable in

the financial performance of the domestic acquirers.

Further examination to ascertain the percentage contribution of other significant variables

was undertaken. The beta values show that CIR has a significant beta value of 38.1% in the

model. Other LOANTAS and RISK have insignificant betas of 10.1% and 18.9%

respectively. The Sig. shows that the CIR is significant at p=0.005 while the others are not.

40

Beta values are generally used in finance to examine the risk and thus the expected returns. Similarly, in

statistics, this measure could specifically be applied in measuring the extent of contribution or the effect exuded

by individual variables in an equation. Though used as a measure of systematic risk, it also employed in

equating specific returns relative to the overall regression (Seiler, 2004). The higher the beta value, the more

contribution and thus impact of the variable on the regression result.

156

Thus, in summary, the regression result shows that PROVTAS and CIR are the most

significant variables affecting the financial performance of domestic bank acquirers. This

implies that the bank‟s liquidity, including its financial position, investments strategy and

funding methods proxied by PROVTAS, are critical to its post merger performance. Merged

bank entities with good liquidity level and well thought out investment strategy are more

likely to outperform those banks with poor liquidity level. Controlling bank expenditure

pattern in relation to the bank income, as measured by the CIR, is also essential if the

domestic merged banks are to strive in their financial performance. As indicated by the

significance of CIR, costs should be matched with the income level of the banks to generate

good performance. The efficiency in management of expenses relative to income is thus

important.

6.4.2 Performance Analysis of Cross Border Acquirers

As in the case of domestic acquirers, the initial stage of the financial performance analysis of

cross border acquirers was to determine the extent of correlation among the performance

variables using the Pearson moment correlation coefficient and to test for any

multicollinearity41

in the regression.

Tables 6.8 and 6.9 show the correlation coefficients of the performance variables of the cross

border acquirers. The pre-merger correlation matrix shows significant high correlation of

ROE with CIR and LOANTAS at -0.639 and -0.305 respectively. These depict that the

higher the profitability of the cross border acquirers, the lower the CIR (used as proxy for

cost to income ratio) and the LOANTAS.

The result shown on the Table 6.8 indicates that the profitability variables do not correlate

with pre merger CTSAR of the cross border acquirers. This simply implies that the events

leading to the M&A announcement do not strongly affect the financial performance of the

cross border acquirers. This position however changes after the merger event. The

coefficients of the post cross border acquirers (Table 6.9) indicate that CTSAR has a

significant correlation at -0.425, 0.574, 0.397 and -0.453 respectively with ROE, CIR,

LOANTAS and SIZE of the bank as measured by the total assets.

41

See the Chapter 5 for discussion on multicollinearity tests

157

This indicates that there is a relationship between abnormal returns and thus shareholders

wealth and cost to-income of the acquirers (CIR) as well as the liquidity of the banks

(LOANTAS). However, such increase in CTSAR attracts a reduction in the profitability and

the total assets of the cross border acquirers. In the same vein, profitability (ROE) is related

to the level of CIR, RISK and CTSAR as each significantly correlates negatively with the

Table 6.8 Correlations of the Cumulative Total Standardised Abnormal Returns with Pre

Cross Border Performance Variables

Variable ROE NII EQTASS CIR LOANTAS PROVTAS RISK SIZE

CTSAR

ROE

1.000

.

NII .108 1.000

EQTASS -.185 .568**

1.000

CIR -.639**

-.254* -.028 1.000

LOANTAS -.305**

.009 .191 .346**

1.000

PROVTAS .020 -.172 -.098 -.136 -.037 1.000

RISK .026 .111 -.007 -.167 -.166 -.009 1.000

SIZE

-.117

-.112

-.062

.020

.124

.098

.126

1.000

CTSAR -.105 .009 -.058 .101 .130 -.157 -.132 163

1.000

**. Correlation is significant at the 0.01 level (2-tailed); * Correlation is significant at 0.05 level (2-tailed)

158

ROE at -0.473, -0.469 and -0.425 respectively .The beta values are used in assessing the

contributions of each of the independent variables in predicting the financial performance of

the cross border acquirers. Table 6.10 shows the beta coefficient results and empirically

evaluates the percentage contributions of each of the variables as measured by the level of

their betas.

Table 6.10 shows the beta values and the t-statistics in parentheses of the cross border

profitability variables. Four models were used to examine the contribution of NII, SIZE,

EQTASS, CIR, LOANTAS, RISK, PROVTAS, CTSAR and dummy variables in the

regression. Model 1 contains two variables; NII and SIZE. Model 2 has, in addition,

Table 6.9 Correlation matrix of the Cumulative Total Standardised Abnormal Returns with

Post Cross border Performance Variables

Variables ROE NII EQTASS CIR LOANTAS PROVTAS RISK SIZE CTSAR

ROE 1.000

NII .061 1.000

EQTASS .061 1.000 1.000

CIR -.473

** -.287 -.287 1.000

LOANTAS -.302 .142 .142 .476

** 1.000

PROVTAS -.069 .064 .064 .067 .377

* 1.000

RISK -.469

** -.176 -.176 .084 .078 .413

* 1.000

SIZE -.076 -.113 -.113 -.362

* -.455

** -.012 .523

** 1.000

CTSAR

-.425* .024 .024 .574

** .397

* -.238 -.146 -.453

* 1.000

**Significant at 0.01 level (2-tailed), *Significant at 0.05 level (2-tailed).

159

EQTASS and Model 3 has three more variables; CIR, LOANTAS and RISK. The final

Model has PROVTAS, CSTAR and DUMMY in addition to all the previous variables.

Whilst the Adjusted R Square measures the contribution of the models, the R Square

Change estimates the impact of each of the variables in the model. The variables in model 1

make a 33.9% impact on the model result while the additional variable in model 2 which is

EQTASS makes a 10.5% contribution to the model result. The additional 3 variables; CIR,

LOANTAS and RISK make a total of 8.5% in model 3. The final model contains in addition,

PROVTAS, CTSAR and DUMMY made a 15.2% contribution to the regression result.

The significance of each of the variables is determined by their beta values shown in the

coefficient table.

This result implies that the models are all significant and thus highly influence the financial

performance of the cross border acquirers. However, it is important to further evaluate the

extent of contribution in the variation of each of the variables by isolating their beta values.

Not that the variable with the highest beta value makes the most significant impact to the

model result. The interpretation of the result is based on the model 4, which is the

summation of the models result.

160

Table 6.10 Model Summary of Hierarchical Regression of Cross border Acquirers

The variables were introduced in a sequential order. The Model summary shows profitability variable of the cross border acquirers, change statistics and ANOVA. The R Square shows that Models 1 and 2 make 33.9% and 44.4% contribution respectively to the dependent variables while Models 3 and 4 explain 45.8% and 51% respectively of the variance in the dependent variable (ROE). These contributions are very significant as depicted by their Sig. ANOVA values.

Variables Model 1 Model 2 Model 3 Model 4 Collinearity

Statistics

Tolerance VIF

NII 0.084 0.286** 0.287** 0.212* 0.603 1.657

(0.981)# (3.096) (3.034) (2.240)

CIR -0.553** -0.561** -0.521** -0.618** 0.621 1.611

(-6.437) (-7.075) (-5.713) (-6.634)

EQTASS -0.382** -0.356** -0.387** 0.614 1.628

(-4.246) (-3.676) (-4.132)

LOANTAS -0.073 -0.141 0.677 1.477

(-0.826) (-1.582)

PROVTAS 0.044 0.082 0.825 1.212

(0.528) (1.011)

RISK -0.092 -0.059 0.939 1.065

(-1.174) (-0.783)

SIZE (0.222)* 0.77 1.298

(-2.685)

CSTAR -0.217* 0.825 1.287

(-3.012)

DUMMY -0.139

(-1.647) 0.758 1.319

Constant 7.656 9.077 7.968 7.432

Mean Sq 1126.77 982.7 507.49 423.45 F-Value 24.88 25.51 13.11 11.83 Sig. ANOVA 0.000** 0.000** 0.000** 0.000** Change Statistics

R Square 0.339 0.444 0.458 0.51 R Sq Change 0.339 0.105 0.085 0.052 Adj. R Sq. 0.325 0.426 0.423 0.467

Std. Error 6.729 6.206 6.222 5.983

Sig.F Change 0.000** 0.000** 0.479 0.011*

Df1 2 1 3 3

Df2 97 96 93 91

# Figures in parentheses are t-values,* significant at 0.05 level, **significant at 0.01 level

161

The beta values of the variables depict their level of predictions. Thus, Model 4 shows that

CIR has the highest beta value of 0.618. This is followed by EQTASS with beta of 0.318;

SIZE with beta of 0.224; CTSAR with beta of 0.217 and NII with a beta of 0.212

respectively.

The significance of the models is corroborated by the ANOVA which shows the overall

significance of the models. Each model in the ANOVA is significant. This is an indication

that each of the performance models of the cross border banks has a significant level of effect

on its profitability. However, the level of contribution or significance of each variable is

measured by the beta value in the coefficient table.

This significant level shows that in the order of predictability, CIR followed by EQTASS is

the strongest predictors of financial performance of the cross border bank acquirers. The

relative contributions and significance of their betas (0.618 and 0.387 respectively) in Table

6.10 which are confirmed by the sig. value of 0.000. This result indicates that controlling

expenditure pattern in relation to the banks income, as measured by the CIR, is essential if

the cross border merged banks are to improve their financial performance (same result was

also obtained in the case of the domestic acquirers). As indicated by the significance of CIR,

costs should be controlled at the level of the banks earnings, if good financial performance is

to be achieved. Management must strive to be efficient in their control of expenses in the

merged entity.

Profitability will also be achieved when the banks are able to use its capital effectively. The

significance of EQTASS is an indication that profits‟ arising from the quality of the banks‟

assets is an important variable in enhancing the performance of the cross border acquired

banks. This finding leans support to Goddard, Molyneux & Wilson (2004a) which assert that

an increase in capital-to assets ratio (CAR) will subsequently increase the profitability of the

banks.

Interestingly, CTSAR also impacts significantly on the financial performance of the cross

border acquirers. This implies that the abnormal returns consequent on the merger

announcement affects the financial performance of the merged entity. The significant of the

variable highlights the sensitivity of the news of the mergers to the operations and financial

results of the banks involved in any M&A.

162

6.4.3 Summary of Analyses on the Financial Performance

The empirical analysis shows that CTSAR of the cross border acquirers rose to above 20%

after the announcement date but fell to its lowest of -40% on event day. It does not show any

substantial rise thereafter but maintained a steady trend of between -20% and -38%. The

domestic acquirers show a more volatile trend in their CTSAR. The CTSAR rose fast after

the announcement to about 10% but fell to about -20% on the event date. The fall is followed

by a fast recovery to about 5% before plummeting downwards to about -23% within the event

period. This indicates that the cross border acquirers have higher as well as steep returns for

their shareholders than domestic acquirers.

The cross border post merger profitability variables (ROE, CIR, LOANTAS, SIZE) show

more significant correlation with CTSAR compared with the few domestic post merger

profitability variables (EQTASS and LOANTAS ) which also significantly correlate with the

CTSAR. The cross border acquirers also show greater significance with profitability

measures as well as more significant correlation with CTSAR than the domestic acquirers.

The highest abnormal returns occurred in the 1st quartile for both acquirers while the 4

th and

3rd

quartiles give the most negative returns for the post cross border and domestic

respectively.

The CIR, with a beta of 38.1% makes the strongest impact to the financial performance of the

domestic merged entities. The fact that the CIR appears significant in both the cross border

and domestic acquirers indicates that management of costs remains a significant variable in

the financial performance particularly after M&A which reflect the investors‟ expectations.

Thus, the regression result shows that cross border acquirers on aggregate, show higher

profitability than their domestic counterparts. CIR, SIZE and the EQTASS are the most

significant variable affecting financial performance of merged banks. In addition, the

CTSAR of the cross border acquirers significantly affect the financial performance of the

merged banks whilst the domestic CTSAR has no significance on the banks performance.

Thus, the null hypothesis (H2) that the cross border acquirers have better financial

performance and show more significant correlation with the CTSAR, than their domestic

counterpart is thereby accepted.

163

Chapter Seven

Factors Influencing Bank Dividend Policy and the M&A Effect

7.0 Introduction

The purpose of this chapter is to apply the results of the cumulative total standardized

abnormal returns (CTSAR) of the cross border and domestic M&A in measuring the impact of

shareholders perception on the dividend policy of bank acquirers. The dividend yield and

payout are used as proxies for dividend policy of the acquirer banks. Other variables such as

agency costs, bank liquidity, profitability, growth & investments, beta (risk), taxes, debt &

financial structures, ownership, size and geographical factors affecting dividend policy are also

analysed. The results of the summary statistics are presented into different panels while the

regression results are based on different specification of the models.

7.1 The Descriptive Analysis

The dividend payout ratio represents the proportion of the earnings that the banks pay out to

shareholders in the form of dividends. It is usually the net profit after taxation and after the

preference dividends announced during the period; expressed as a percentage. The dividend

yield relates the cash return from a share to its current market value. This is often stated at the

grossed up rate.

The descriptive statistics on the measures of dividend policy42

of the sample commercial banks

are presented in Table 7.1 below. The table compares the key dividend policy measures

42

The concept of dividend policy has been dealt with in the chapter 2. It is important to re-emphasize that in

finance literature, different dividend policy measures have been applied depending on the perspective of the

researchers. However, three measures; yield, payout and EPS have commonly been used in part or jointly to

describe dividend policy. Chen et al (2005) used payout and yield; Gugler and Yurtoglu (2003) used payout;

while La- Porta et al (2000) used series of dividend ratios such as dividend to sales, dividend to earning etc as

instruments of measuring a firm‟s dividend policy. This study uses the dividend yield and dividend payouts to

measure dividend policy, as these are the most commonly used.

164

(dividend payout and yield) of the pre and post mergers of both the domestic and the cross

border sample of commercial banks during the period 1997 to 2007. The table shows a decline

in the two dividend policy measures of dividend payout and yield in the cross border merged

banks. For instance, the mean of the dividend payout fell from 6.63% to 5.44% while the

dividend yield also fell from 0.59% to 0.23% after the mergers.

This result suggests that cross border banks are often under difficulty to increase dividend after

mergers. The length of time to consolidate and recapitalise the banks in an effort to gain

market share may be factors for this decline in the dividend policy. Foreign acquirers would

usually require building a solid capital base in a new market to compete favourably. This may

account for the reduction in the returns and earnings available to the shareholders.

The results for the cross border sample are contrasted with the results of the domestic

acquirers. The means of the dividend policy measures indicate an increase in both the

dividend payout and yield. The dividend payout rose from 7.15% to 8.33% and the dividend

yield also grew from 0.05% to 0.80% respectively.

Table 7.1 Descriptive Statistics on Measures of Dividend Policy for Domestic & Cross border

M&A (1997-2007)

Cross border M&A

Observations

Minimum

Maximum

Mean

Standard Dev

Pre-mergers

Dividend payout % 77 0.04 56.00 6.63 13.03

Dividend yield 64 0.00 5.11 0.59 1.22

Post-mergers

Dividend payout % 79 0.03 25.32 2.87 5.44

Dividend yield 72 0.01 3.29 0.23 0.58

Domestic M&A

Pre-mergers Observations Minimum Maximum Mean Standard Dev

Dividend payout % 57 0.05 86.60 7.15 13.59

Dividend yield 22 0.02 0.12 0.05 0.03

Post-mergers

Dividend payout % 43 0.00 58 8.33 15.81

Dividend yield 24 0.00 3.91 0.80 1.15

165

The rise in the dividend payout and yield in the case of domestic bank M&A may be a bold

attempt to consolidate in the domestic market. Since dividend has a signalling effect, a rise will

be an indication of positive growth and a reassurance to the shareholders43

. The trend in payout

and yield over time are often used to evaluate the potential of a business. Atrill & McLaney

(2004) assert that increase in profit through shareholders investments does not necessarily imply

that the dividend yield will rise as a result.

Dividend policy cannot be considered in isolation. Other factors affecting dividend policy must

be considered in tandem to evaluate the effect of merged entities on dividend policy. Table 7.2

gives a summary of the merged entities‟ dividend policy and the various factors44

affecting it.

It is important to note that the coefficients of variations45

(CV) in the two dividend policy

measures are similar though their means differ. While the CV of the dividend payout is 0.426

the yield shows a variation of 0.551. This implies that the probability differences of the mean of

the variables from their standard deviation lie on the same range. In addition to the dummy

variables, the PE ratio (PER), FNST and the DebtEquity also show same pattern in their

coefficient of variations of 0.496, 0.58 and 0.493 respectively. The beta variable has a mean of

1.108 and a CV of 0.213.

43

The signalling effect of dividend was discussed in Chapters 2 & 3 on the review of dividend policy literature.

However, some studies have revealed that increase in dividend does not necessarily indicate better performance

of the firm. Management often sticks to a pattern of dividend payout. This can be explained by the signal

conveyed by an alteration in dividend payment of the company particularly for new foreign banks. A cut in

dividends will give a negative signal to investors on the future prospect of the company (Hutchinson, 1995).

This might affect the share price of the company. Thus Chang and Rhee (1990) added that financial leverage is a

crucial factor in dividend policy of firms. A firm that has a high financial leverage tends to have a high dividend

payout ratio.

44 The factors affecting dividend policy are quite huge in finance literature. These include cash flow; voting

rights of shareholders, share ownership, Tobin‟s Q, debt ratio, corporate size (Gulger & Yurtoglu, 2003).

Others are civil laws, dividend tax advantage (La Porta et al, 2000), profitability, ownership structure,

investment and growth opportunities, and the assets structure of the firm (Myers and Majluf, 1984; Rozeff,

1982; Jensen et al 1992; and Barclay et al, 1995 and Fama & French, 2000). Baker & Powell (2000) also

classified them into economic and industry specific factors, McManus et al (2003) add beta (risk) factor while

Anil & Kapoor (2008) consider earnings (current and future), liquidity, corporate tax. See chapter 2 for

discussion on factors affecting dividend policy.

45 The Coefficient of Variation (CV) is computed as standard deviation/mean

166

Ta Table 7.2 Univariate Statistics of Measures and Factors Affecting Dividend Policy

The table presents the results of the univariate statistics of factors affecting dividend policy. The variables are

presented in different panels. These include dividend , profitability, CTSAR, risk, taxes, debts & financial

S structure, growth & investment, size, ownership and liquidity variables.

Variables Variable Code Minimum Maximum Mean Standard Deviation

Coefficient of Variation

Panel A: Dividend Policy Variables

Dividend Payout ratio Payout 5.79 100 50.78 21.663 0.426

Net Dividend Yield Yield 2.3 27.25 9.015 4.968 0.551

Panel B: Profitability

Earnings Per Share EPS -4.26 41.2 3.224 4.929 1.528

Return on Equity ROE -42.8 45.92 10.942 9.489 0.867

Panel C: Cumulative total standardized abnormal returns CTSAR -48.14 29.02 -11.274 15.986 -1.418

Panel D: Risk

Beta Beta 0.6 1.55 1.108 0.236 0.213

Panel E: Taxes

Tax Tax 0.01 8.62 1.108 1.3 1.173

Pre-Tax Operation/ Average Assets Pre -5.56 67 1.055 5.207 4.935

Non operation item & Taxes/Average Assets

Nonoptax/ Ass -2.36 1.6 -0.134 0.389 -2.908

Panel F: Debt & Financial Structure

Debts + equity/total assets FNST 0.07 15.14 5.289 3.072 0.58

Total Debt/Equity DebtEquity 1.34 16.17 5.761 2.842 0.493

Panel G: Growth & Investment

Price Earning Ratio PER 0.35 15.96 5.958 2.956 0.496

Market to Book Value Ratio MBR 0.06 40.79 3.94 9.937 2.522

Panel H: Size

Nat. Log of Total Assets NLTOTASS 0.01 7.43 0.42 0.98 2.33

Nat. Log Total Cap. Capitalisation 0.03 10.37 1.999 2.848 1.424

Panel I: Ownership

% Insider holdings INSIDER 0.07 68.5 30.116 21.617 0.717

Panel J: Liquidity

Net Cash/Total Assets NetcashASS -39.88 62.19 -0.14 8.034 -57.066

167

The CTSAR shows a negative mean of -11.274 and a CV of -1.418. This result implies that

the banks CTSAR make no significant contributions to the dividend received by the

shareholders from the merger activities. To verify the assumption, the results are subjected to

further analyses.

7.2 Test of the Hypothesis of the Effect of M&A on Dividend Policy

H3: The dividend policy hypothesis: There is no significant difference in the shareholders

perception of dividend policy of cross border and domestic bank acquirers following a

merger or acquisition.

There are no known studies which have focused on the impact of M&A on the dividend

policy of banks, although there are many studies that have devoted enormous strength in

researching on different aspects of M&A as well as dividend policy.

The hypothesis was formulated using key dividend variables from known studies (Rozeff,

1982; Casey & Dickens, 2000; Chang & Rhee, 1990; DeAngelo, DeAngelo & Stilz, 2006;

Chen et al, 2005) and applying CTSAR as a proxy for bank M&A effect. The dividend yield

and payout are two variables commonly used in the finance literature as measures of dividend

policy 46

.

The regression model on the effect of M&A on dividend policy has the following

specifications below and the results are presented in Tables 7.3.

46

See Chapters 2 & 3 for the literature review on dividend policy for discussion on the measures of dividend

policy.

168

DivPolicy = - Betai + Liquidityi - Insideri + Taxi - Capst + β7FnSti + β 8Sizei

+ β9Profiti - β10Growthi + β 11CTSARi + ξi

Where the effect of each of each explanatory variable is positive except where otherwise

indicated:

- intercept term;

Beta - Estimated beta coefficient of the banks (with negative sign to indicate its expected

effect)

Liquidity – the availability of physical cash in the bank measured as the dividend/net cash

operating,

Insider – the percentage of insider shareholdings in the acquiring banks (with expected

effect being negative);

Tax – the total tax liabilities of the banks as well as the relevant tax ratios;

CapSt - bank‟s capital and finance structures measured by the Debt/Equity ratio.

β6FnSt – the finance structure is debt + equity/total assets;

β7Size – the natural log of the total assets is used as the proxy for size of the bank;

β8Profit – the profitability of the bank as measured by the ROE and EPS;

β9Growth – the price earnings (PE), which also represents the market to book ratio (MBR)

which is a proxy for Tobin‟s Q measure future growth and investment of the bank (with

negative expected effect on dividend policy);

β10CTSAR – the Cumulative total standardized abnormal returns (CTSAR) is a proxy for

M&A;

ξ - error term

169

Table 7.3 Model Summary of Hierarchical Regression of Dividend Policy

The two dependent variables (DVs); dividend payout and dividend yield, were used to regress against a

number of dividend policy factors, and the using the CTSAR as the independent variable (IV). The IVs were

introduced in sequence to identify 5 different models. The figures in parentheses are for the values obtained

using the yield as the dependent variable while the others are for the values obtained using the payout as the

dependent variable.

Variables Model 1 Model 2 Model 3 Model 4 Model 5

Insider 0.033 0.035 0.043 0.029 0.008

(-0.017) (-0.077) (-0.088) (-0.098) (-0.131)

Liquidity -0.087 -0.090 -0.098 -0.132 -0.09

(0.299) (0.383)** (0.362)** (0.325)* (0.325)*

Beta -0.131 -0.145 -0.149 -0.158 -0.143

(0.452)** (0.491)** (0.523)** (0.519)** (0.538)**

Tax 0.055 0.057 0.036 0.04

(-0.304)* (-0.275)* (-0.291)* (-0.308)*

TaxToTass -0.039 -0.041 -0.02 -0.062

(0.328)* (0.314)* (0.337)* (0.377)**

NonTAX -0.03 -0.024 -0.004 -0.019

(-0.084) (-0.053) (-0.028) (-0.010)

FnSt -0.028 -0.042 -0.044

(-0.324)* (-0.328)* (-0.370)**

TotalAss 0.055 0.075 0.04

(-0.017) (0.003) (-0.009)

CapSt 0.005 0.034 -0.045

(0.001) (0.025) (0.060)

EPS 0.07 0.035

(0.126) (0.155)

ROE -0.164 -0.144

(-0.159) (-0.170)

PE -0.054 -0.046

(-0.046) (-0.056)

CTSAR 0.097

(0.250)

Constant 1.372 1.336 1.191 1.281 1.245

(1.277)# (1.539) (2.013) (2.003) (2.093)*

Mean Sq. 10.295 6.059 4.513 5.914 7.315

(17.525) (15.501) (12.349) (9.798) (9.266)

F-Value 0.429 0.237 0.164 0.204 0.246

(4.633) (5.449) (4.937) (3.851) (3.662)

Sig. ANOVA 0.733 0.962 0.997 0.997 0.996

(0.008)** (0.001)** (0.001)** (0.002)** (0.003)**

Change Statistics R Square 0.028 0.033 0.037 0.065 0.094

(0.290) (0.513) (0.613) (0.649) (0.665)

R Sq Change 0.028 0.005 0.004 0.028 0.029

(0.290) (0.223) (0.100) (0.035) (0.016)

Adj. R Sq. -0.038 -0.108 -0.191 -0.255 -0.290

(0.228) (0.419) (0.489) (0.480) (0.483)

Std. Error 4.896 5.059 5.244 5.384 5.458

(1.945) (1.687) (1.582) (1.595) (1.591)

Sig. F Change 0.733 0.975 0.984 0.790 0.595

(0.008)** (0.008)** (0.087) (0.484) (0.267)

Df1 3 3 3 3 1

Df2 144 141 138 135 134

*Significant at 0.05 level, **Significant at 0.01 level. #Values in parentheses are for results obtained using dividend yield.

170

Table 7.3 shows the Summary Model of the hierarchical regression models and compares the

results of both the payout and yield. The yield results are in bracket which indicates the

extent of significant impact of the regression model. The results of the regression are

presented in 5 different models, by including additional variables into the successive models.

This serves as a robustness check by evaluating the specific impact of any variable added to

the model.

The Model Summary table shows that at a glance, the R values of the dividend yields are

quite higher than the dividend payout. The R value shows the linearity in the correlation

between the observed and the model predicted values of the dependent variables. A larger

size will indicate a stronger relationship. Thus the independent variables show a higher effect

with dividend yield than they do with the dividend payout.

This position is further highlighted by the R Square, which shows that more than half of the

variation of the dividend yield is explained by the model. For instance, the R value of the

dividend yield is 0.665 (the Model 5 of the Model Summary table). This indicates the strength

of the model fit and is corroborated by the low Standard Error of the Estimate of the dividend

yield which is below 2.

The discussion of the results follows a sequential order by examining each of the independent

variables in the models. Since CTSAR is in Model 5, which is the proxy for M&A and the

variable of main interest to the hypothesis, the discussion of the results begins with that

model.

7.3 The Effect of CTSAR on the Dividend Policy

The behaviour of the abnormal returns as hypothesized in the study is a reflection of

investors‟ expectations about dividends. Thus, the accuracy of such expectations is tested by

the dividend regression results. Dividend policy here is measured by the dividend yield rather

than dividend payout47

. The Model 5 results show that the introduction of CTSAR in the

47

Dividend payout as a measure of dividend policy is excluded from discussion here because most of the

explanatory variables are insignificance in the regression models. The Coefficient table (Table 7.5) also shows

171

regression had a slight improvement on the strength of the relationship and accounts for 25%

of the variance in the model. The R Square barely changed from 64.9% to 66.5%. The Sig. F

Change shows that the contribution is insignificant at P= 0.267. Specifically, the CTSAR

makes just a 0.003% impact on the model as shown by the R Square Change after

controlling for the effect of the other variables.

However, the ANOVA result in shows that Model 5 as a whole is very significant at

P=0.003. This shows the acceptability of the model from a statistical perspective, though it

does not directly address the strength of the relationship of each of the individual independent

variables. This relationship is determined by the beta coefficient values.

Since model 5 makes a significant contribution to the regression, it is important to isolate the

impact of the individual variables particularly the CTSAR. The last model48

summarises the

contributions of each variable. The Beta of the standardized coefficients of CTSAR is 0.250.

This indicates, however, that the contribution49

of CTSAR in the overall model development

is insignificant and the null hypothesis that shareholder perfection from M&A (which uses

CTSAR as a proxy) has no significant effect on the dividend policy (measured by the

dividend yield) of banks is therefore accepted.

7.3.1. Robustness check of the Merger Effect on Dividend Policy

The Granger Causality effect is applied to check more vigorously the existence of any temporal cause

and effect relationship between two variables (Seiler, 2003); in this case, the dividend yield and the

CTSAR. Therefore, a Granger Causality model is employed to check the presence of any

that several of the dividend payout predictors are not significant, thus indicating that the variables do not

contribute much to the model.

48 Model 6 is a summary of the results of the variables used in the regression. It specifies the contribution and

thus the significance of the predictor variables. Thus, the analyses of the variables are based on the data in

Model 6.

49 The Beta of the Standardized Coefficients offers an indication of the relative strength and contribution of the

predictor variable to the model. A high Beta value indicates that the variable exerts more influence on the

model. Beta values of at least to 0.300 generally make significant contributions (Pallant, 2004).

172

causality relationship between the CTSAR variable and the dividend policy. The result is

shown in Table 7. 4.

Table 7.4 Granger Causality Test Result

Pairwise Granger Causality Tests

Samples: 1/1/1997 – 31/12/2007

Lags: 5

Null Hypothesis: Obs F-Statistics Probability

CTSAR does not Granger Cause Dividend yield 155 1.9650 0.1231

The p-value (probability) of the result shows a value of 0.1231 which is not significant. In

statistical terms, this implies that the banks‟ dividend policy is not significantly affected by

the share price reaction from the news of mergers and acquisitions.

This result concurs with the results of the regression model. Thus, the null hypothesis that

announcement effects of mergers & Acquisitions have no impact on dividend policy is

accepted.

The rest of the discussion focuses on the other dividend variables used in the models; to

ascertain the extent of their individual impact on the dividend policy of the banks.

7.3.2. The Impact of Insider Shareholdings (Agency Costs), Liquidity and Beta on

Dividend Policy

The result of Model Summary in Table 7.3 indicates that the insider shareholdings percentage

(which is a proxy for the agency costs) is an important variable in the determination of the

dividend policy. The Model 1 result (which contains in addition to the Insiders variable,

Liquidity and Beta variables) has an R Square Change of 29% and significant p-value of

0.008. This signifies that the combined set of variables in the model exert a significance

173

impact on the dividend policy of the banks. This position is further supported by the ANOVA

result which shows that the model shows a significant p-value of 0.008.

However, to obtain the actual contribution of the agency costs variable. The variable Insider

shows a standardized beta coefficient value of -0.131. This is insignificant and shows that

although the overall Model has a significant impact (ANOVA of 0.003) on dividend policy,

the Insider variable does not make any significant contribution to the model.

Although the initial expectation predicted by the model is that the Insider would make a

significant impact on the dividend policy, however the result obtained is totally not

surprising. Finance literatures50

have been agog on the issue of agency costs and its impact on

dividend policy of firm. Rozeff (1982) found that agency costs (Insider) and beta have no

significant effect on the dividend of banks. This result corroborates Casey and Dickey (2000)

who also found that Insider makes no impact to the dividend policy of banks. Others, such as

Baker, Gary & Powell (2000) and Johnson, Lin & Song (2006) also confirm that the

percentage of Insider has no bearing on the dividend policy of firms. Dempsey & Laber

(1992) add that while the dividend yield is negatively related to the Insiders, it has a positive

significance to the proportion of the ordinary shareholders.

However, there are also quite a number of opposite findings asserting that agency cost is a

significant element on the dividend policy of firms. Results obtained by Wei, Zhang & Xiao

(2003) and Gugler (2003) indicate that the agency costs is an influential variable in dividend

policy considerations. Brunarski, Harman & Kehr (2000) assert that a higher number of

insider helps to curtail the wasteful spending and channel funds towards positive projects and

thus determine greatly the dividend policy. Additionally, a recent study by Cornett, et al

(2008) affirms that the number of insiders is an important variable in the dividend decisions

of public banks.

It is therefore apparent that the extent of the insider influence on dividend policy decisions is

inconclusive as well as debatable. Any position taken must be justified in conjunction with

other variables.

50

Many studies on the role of insiders on dividend policy have been undertaken over the past three

decades. Short (1994) and Gugler (2003) present an extensive survey of studies in dividend policy

cum agency costs.

174

The ANOVA and the Model Summary results also confirm that at P=0.008, the Liquidity51

variable in Model 1 is a significant factor in the dividend policy of banks. The strength of its

contribution is denoted by its beta coefficient of 32.5. Model 5 summarises the contributions

of each of the variables, shows that liquidity makes a significant contribution to the dividend

policy decisions of the banks.

The result obtained is consistent with the majority of findings on the impact of liquidity or

cash flow on dividend policy. Lie (2000) asserts that cash dividend declaration is positively

related to the firm‟s level of liquidity. Gugler (2003) attributes the liquidity of a firm as very

fundamental in its dividend payout. While La Porta, et al (2003) affirm that the cash position

of the firm is very crucial in its dividend decision.

A recent study by Anil & Kahoor (2008) also confirm that good liquidity position increases a

firm‟s ability to pay dividend as those firms with unstable cash flows are less likely to have a

regular dividend. The aforementioned studies thus confirm that the result on the liquidity is

in conformity with previous studies.

The table also shows that the Beta variable, which is a proxy for risk, is a very significant

factor in the dividend policy of bank acquirers. It has a significant standardized beta

coefficient of 0.538. The result finds support from previous studies such as Rozeff (1982),

Blume (1980) and Massa & Zhang (2009) all of which found the effect of risk represented by

beta very significant in dividend policy.

Pang, et al (2008) posit that the dividend yield always follows the pattern of the beta and

employ the coefficient of variation to measure the stability of the yield. This procedure

highlights the importance of the beta variable. Pandey (2001) using 1729 Malaysian firms in

a panel data analysis found that beta is significant in measuring the dividend yield. Watson

& Head (2004) affirmed that firms such as banks that operate in high business risk ventures,

which are susceptible to cyclical swings in profit, tend to respond by paying low dividends in

order to avoid the risk of reducing dividend in the future.

51

The liquidity is measured as the dividend / net cash operating. It denotes the cash available after all capital

expenditures have been undertaken before payment of ordinary dividend. The ratio does not take into account

stock dividend payments as those do not require cash and previous period under/over provision payments.

175

However, the issue of the role of beta in dividend is no less contentious in finance studies.

There have been a number of opposite findings. Casey and Dickens (2000)52

found no

significance in the role of beta in dividend policy of banks and challenged the earlier findings

of Rozeff (1982). In the same vein, Chen, Grundy & Stambaugh (1990) investigated the cross

sectional relationship between the dividend yield and market risk (beta) using the market and

changing risk premium approaches and found that none has a significant effect on the other in

both methods. Despite the above opposing views, most contemporary studies are in

agreement that beta is an important variable in dividend decisions.

7.3.3. The Impact of Taxes on Dividend Policy

Apart from the actual tax paid, two additional tax variables are included to provide a robust

check53

on the actual tax effect on dividend policy. The additional variables include the Pre-

tax operation income to Average assets (TaxToTass) and the Non operational items & taxes

to Average assets (NonTAX).

The Model Summary table shows that after controlling for the effect of the previously

considered variables (insider, liquidity and beta), taxes are very significant at P=0.008. This

is further confirmed by the ANOVA table of P=0.001. To determine the strength of each of

the tax variables in the Model 2, the beta coefficient shows that the actual Tax and

TaxToTass liabilities of the banks all make a significant input -0.308 and 0.307 respectively.

The NonTAX variable showed no significance. The result is not surprising as taxes generally

influence the dividend payout of firm, hence dividend are usually paid from profit after tax

(PAT).

52

Much of Casey & Dickens (2000) findings was a cross examination of the earlier study by Rozeff. They used

similar variables as Rozeff and found differences in the results. Three outstanding variables were particularly of

interest in their findings (the firm‟s growth rate, insider, and beta) all of which were insignificant and opposite

of Rozeff‟s findings.

53 Both the TaxToTass and NonTAX were incorporated to assess the operational and non operational impacts of

incomes before the provision and payment of taxes by the banks. It is expected that the 2 variables will balance

off any differences resulting from the fluctuations in the tax regimes.

176

However, previous studies relating to the tax effect on dividend decisions have produced very

conflicting results54

. Casey & Dickens (2000) affirmed that taxes have significant impact on

the dividends of commercial banks in the US. Their findings concurred with an earlier study

by Rozeff (1980). The assumption is that the lower the taxes, the higher the dividend payout.

In addition to the dissident findings, Anil & Kapoor (2008) maintain that the imposition of

taxes on dividend has no significant impact on the dividend policy of any organisation.

Wu (1996) investigated the impact of eliminating the preferential capital gain tax treatment of

1986 in the US and found some structural changes in the pattern of dividend which coincides

with changes in the tax laws. The study concludes that such a shift significantly affects the

aggregate corporate dividend payout. Wilkinson, Cahan & Jones (2001) recommended a

reduced tax policy for firms in New Zealand as a strategy for dividend imputation. In a

recent study, Pattenden & Twite (2008) evaluated the tax effect on dividend policy in

Australia under different tax regimes for the period 1982-1997. They found that the increase

in dividend payout and initiation differ among different firms. However, the study affirms

that the higher the level of available franking tax credits, the higher the dividend initiation

and payout.

54

La Porta, et al (2000) & Poterba & Summers (1985) chronicled various studies and highlight the various

divergent views among economists of the tax effect on dividend policy. The traditional views assert that high

taxes (either on personal or corporate bases) particularly in the US often serve as a bulwark to dividend

payments. But this position is not without objections. Miller & Scholes (1978) held that investors employ

various dividend tax avoidance techniques that make them escape from taxes. The „‟new view of dividends and

taxes‟‟ proponents such as Harris, Hubbard, and Kemsley (1997), assert that taxes do not deter dividend

payments. They agreed that cash must be paid out as dividend to shareholders at some point so, the payment of

such dividends imposes no great burden on the shareholders.

177

7.3.4. The Impact of Capital & Finance Structures and Size on Dividend Policy

The capital structure (CapSt) generally refers to the composition of the debt and equity in the

banks‟ financial statements. The variable consists of the total debts/ total shareholders‟ equity

(DebtEquity) ratio55

. Model 3 contains in addition to the capital structure, the finance

structure and size variables. The Model Summary shows that all three variables combined

make an insignificant impact on the dependent variable. The R Square Change of Model 3

makes 0.100 (10%) of the regression total model.

However, the ANOVA table shows that all the 3 variables (debt/equity structure, capital

structure and size), while controlling for the earlier variables in the Models 1 and 2, have a

significant value of 0.001. The beta coefficient provides a clue on the individual significance

of each of the variables and shows that the CapSt has a standardized coefficient beta of 0.060

which is insignificant. This means that the level of debt/equity does not influence dividend

policy; the higher the debt, the higher the dividend yield.

The result concurs with the findings of Baker & Powell (2000) and Brunarski, Harman &

Kehr (2004) that the leverage level of firms has no significance on the level of their

dividends. Their argument is based on the principle of the information content of dividend

which opines that a reduction in the dividend may send a signal of cash flow problems to the

public.

The debt structure of firms has constituted a platform in finance for divergent arguments. Not

all agree to the non significance of a firm‟s leverage on dividend policy. Gugler (2003) and

Pattenden & Twite (2008) affirm that leverage greatly influences the dividend pattern of the

firms. The composition of the debt structure is thus very important. Due to the cost of raising

equity, banks that have higher proportion of external financing than equity capital are likely

to pay more dividends.

In fact, Kanatas & Qi (2004) posit that the firms‟ debt financing can be greater or less than

the funding needs if the difference is paid to shareholders. A low debt structure increases the

55

The ratio fundamentally provides a fundamental test of the financial strength of a company. Its purpose is to

create a picture of the measure of the mix of funds in the balance sheet and to make a comparison between those

funds that have been borrowed (debts) and those that have been supplied by the owners (equity). As debts cost

less than the equity funds, the inclusion of debts could improve profitability, add value to the share price,

increase the wealth of its shareholders and develop greater potential for growth (Walsh, 2008).

178

firm‟s value and increases the cash available for disbursement as dividend. Conversely, a

default in meeting up the debt obligations of the firm may trigger the suspicion of their

lenders to inspect the firm‟s debt covenants; any serious violation on the part of the firm may

jeopardise its going concern.

The finance structure of the banks as measured by the (debt + equity) / total assets (FnSt)56

,

makes a significant impact. This implies that the banks‟ finance structure is significantly

important in the dividend decisions. The variable has a beta coefficient of -0.370, which is

statistically significant.

Most past studies on the role of finance structure and dividend lean support to this finding.

Brunarski, Harman & Kehr (2004) and Pattenden & Twite (2008) investigated the optimal

finance structure of firms and assert that the assets and equity composition of the finance and

capital structures as well as its fixed and current proportions are very important in its

decisions on dividend. When the equity/ asset ratio increases, the dividend decisions will be

reviewed upwards. As the number of shareholders increase, their stake also increase in the

organisation, thus this affects the review of the dividend policy of the banks.

The TotalAss, which is used as proxy for size, shows a non significant value and a

standardized beta coefficient value of -0.009. This indicates an insignificant contribution and

points out that size of a bank is not a sinquanon in its dividend policy. Chang & Rhee (2003)

and Johnson et al (2006) argued in support, that the size does has no bearing on the dividend.

Pattenden & Twite (2008) found the same result and observed that large firms, with their high

level of debts, do not necessarily pay better dividend. Also firms with many high equity

capitals do not guarantee a higher dividend policy. The result however is inconsistent with

the findings of Reeding (1997) and Fama & French (2001) that large firms are likely to pay

dividends.

56

This is often confused with capital structure. It refers to the financing of the firm‟s assets based on the totals

of the short-term borrowing, long-term debts and owner‟s equity. The capital structure is primarily focused on

the long-term debt cum assets.

179

7.3.5. The Impact of Profitability and Growth on Dividend Policy

Two variables, ROE and EPS, are used as measures of profitability57

. The PE variable is

used as proxy for growth. The Model Summary Table 7.3 shows that the three variables

combined make no significance difference to the model, given their significance of P=0.484.

The R Square Change is 3.5% while the ANOVA indicates however, that in combination of

the previously considered independent variables, Model 4 turns significant at P=0.002.

Furthermore, Model 5 of the coefficient Table 7 shows that EPS and ROE have beta values

of 0.155 and -0.170 respectively while the PE has a beta value of -0.056. Both the

profitability and growth variables show no significance as dividend policy measures.

The result is contrary to the conventional wisdom, as profit is expected to influence the

dividend decision of the banks and growth reduces the total amount of dividend available to

the shareholders of the banks. However, previous studies have found similar results. In fact,

there are two schools of thought on the influence of profitability on dividend policy. One

opposes the hypothesis that profitability affects dividend while the other supports it.

Studies opposing profitability as a determinant of dividend, and thus giving credence to the

result obtained in the study include Change & Rhee (1990), Baker & Powell (2000) and Anil

& Kapoor (2008). The justification of their argument lies in the fact that a reduction in

dividend due to a decrease in profit gives a bad signal about the bank. Banks would maintain

a sustainable level of dividend such that a downturn in the organisation would not lead to a

reduction in dividend. In fact, these proponents believe that firms would rather increase their

leverage than reduce their dividends.

57

Different measures of profitability have been used in profitability studies. The earnings per share (EPS) relate

the earnings generated by the bank which is available to the shareholders to the number of shares in issue. It is

measured by the after tax profit less any preference dividend divided by the number of ordinary shares. The EPS

measures the absolute return delivered to the shareholders. Growth in the EPS indicates the progress and profit of

the bank.. It is a very powerful indicator of financial performance of a firm (see also studies by Kumar &

Sopariwala, 1992 and Kaufmann, Gordon and Owers, 2000). At the level of the individual firms, the ROE keeps

in place the financial framework for a thriving and growing enterprise and drives industrial investment, growth in

GNP, employment, government tax receipts at the macro economic level (Walsh, 2008). Apart from the ROE and

EPS, the ROCE, returns on net worth and net profit margin are also profitability measures (See Chander and

Priyanka, 2007).

180

Findings supporting profitability as an influential element in dividend policy include Gaver &

Gaver (1993), Fama & French (2001, 2002), Gugler (2003) and Pattenden & Twite (2008).

Azhagaiah & Sabari (2008) also found profitability as an important element in the dividend

decision of Indiana chemical industry. The argument portrayed in these studies is that profit

is directly related to the dividend. Thus, a fall in profitability will amount to a decrease in the

amount of dividends declared and paid, thus a decline in the dividend yield. The argument

does not however take into consideration that a reduction in dividend due to a fall in profit

would send a wrong signal to the public and could thus jeopardise the growth of the bank.

The growth variable58

, PE ratio, is also not significant. It has an insignificant standardised

beta coefficient of -0.059. The negative beta coefficient implies that the banks‟ growth and

dividend are negatively correlated. An increase in the bank‟s growth will decrease the

dividend.

Baker & Powell (2002), Anil & Kapoor (2008), Chang & Rhee (1990), Pattenden & Twite

(2008) and Casey & Dickens (2000) are all in agreement that growth of a firm has no

significance on its dividend policy. The dividend signal hypothesis eliminates any idea of

dividend reduction. Thus, the argument is that when a bank grows, it increases both capital

and finance structures at the same level with its dividend policy.

However, like in many other studies, there are contradictory findings against this view. Some

studies (Gaver & Gaver 2003, Grullon, et al 2002, Fama & French 2002 and Brunarski,

Harman & Kehr 2004) have found that increase in growth would potentially drain the

earnings available to shareholders and thus reduce dividend. Rozeff (1982) and

Gugler (2003) add that the significance is negative as the increase in one causes a reduction

in the other. However, this argument does not lean support to the market reaction to any

significant negative impact on dividend.

58

Different growth proxies that can be used include retained earnings as a percentage of owners‟ equity, current

assets/ sales; retained earnings/sales (Walsh, 2008). But most studies agree that the PE ratio is a credible

measure of growth of a firm. The ratio measures the future earning growth of the firm. A modified ratio, the

PEG also measures the firm‟s growth by dividing the PE by the short term earnings growth rate. Increases in

sales and total assets are also often used to measure growth (Easton, 2004). The argument is that when a bank

grows, it requires capital for expansion. Such funds will thus reduce the available sum set outside for dividend.

181

7.5. Summary of Discussions

Thus in summary, CTSAR does not affect the dividend policy of the both the domestic and

cross border acquirers. The coefficient result shows that it has an insignificant impact on the

dividend policy of the sample banks. This finding is corroborated by the Granger causality

test result, which showed that M&A does not affect the dividend policy of banks.

CSTAR is insignificant on both measures of dividend policy used in the study (dividend yield

and payout). The insignificance of the cross border and domestic CTSAR variable implies

that the dividend policy is not influenced by the geographical locations or national boundaries

of the banks and remains same even after the event. In the same vein, the profitability, capital

structure (debt/equity), size (total assets), growth (PE) do not also influence the dividend

policy of both the domestic and cross border acquirers.

However, other variables such as the beta, liquidity, taxes, equity to total assets and the

finance structure of the acquirers are significant variables in the formulation of the dividend

policy of the banks. These findings confirm the results of previous studies on dividend

policy. The beta, which a proxy for risk, is the most significant factors affecting the dividend

policy of the banks.

182

List of References

Adams, R. & Mehran, H. (2003). Is Corporate Governance Different for Bank

Holding Companies? FRBNY Economic Policy Review, 123-142.

Aggarwal, R. & Samwick, A. (2003). Why Do Managers Diversify their Firms?

Agency Reconsidered. Journal of Finance 58(1), 71-118.

Ahmed, A.S., Takeda, C. & Shawn, T. (1989) Bank loan loss provisions: a Re-

Examination of Capital management, earnings management and signalling

Effects, Journal of Accounting & Finance.

Ahuja, G. & Katila, R. (2001). Technological acquisitions and the innovation

performance of acquiring firms: A longitudinal study. Strategic Management

Journal 22(3), 197 – 227.

Akerlof, G.A. (1970). The Market for 'Lemons': Quality Uncertainty and the Market

Mechanism. The Quarterly Journal of Economics 84 (3), 488-500.

Albert, D. & Pamela, M. (2007). A Longitudinal Study of Net Interest Margin by

Bank Assets Size: 1992-2005 Journal of Economics and Finance, 1995 – 2005.

Albright, S.C., Winston, W.C. & Zappe, C.( 1996). Data Analysis and Decision

Making with Microsoft Excel. Pacific Grove; Duxury Press.

Allen, F., Antonio E. B., and Ivo W. (2000) A Theory of Dividends Based on Tax

Clienteles, Journal of Finance 55, 2499-2536.

Allen, F., and Roni, M. (2002) Payout Policy, (in) George Constantinides, Milton Harris,

And Rene Stulz, eds.: North-Holland Handbooks of Economics, Amsterdam, Elsevier.

Aharony, J., and Itzhak, S. (1980), Quarterly Dividend and Earnings Announcements and

Stockholders' Returns: An Empirical Analysis, Journal of Finance 35, 1-12.

Alli, K. L. A.; Qayyum, K., and Gabriel G. R. (1993), Determinants of Corporate

Dividend Policy: A Factorial Analysis, The Financial Review 28, 523-547.

Altunbas, Y. & Marques. (2008) Mergers & acquisitions and bank performance in

Europe: The role of strategic similarities. Journal of Economics & Business

60,204-222.

Ambarish, R., Kose, J. and Joseph, W. (1987), Efficient Signalling with Dividends and

Investments, Journal of Finance 42, 321-343.

Amihud, Y. and Maurizio, M. (1997), Dividends, Taxes, and Signalling: Evidence from

Germany, Journal of Finance 52, 397-408.

183

Amel, D. and Liang, J. N. (1997). Determinants of Entry and Profits in Local

Banking Markets. Review of Industrial Organization. 12(1) February: 59‐78.

Amihud, Y., DeYoung, G. & Saunders, A. (2002). The Effects of Cross Border Bank

Mergers on Bank Risk and Value. Journal of International Money and

Finance 21, 857 -877.

Andrade, G., Mitchell, M. & Stafford, E. (2001). New Evidence and Perspectives on

Mergers. The Journal of Economic Perspectives 15(2), 103 -120.

Ang, J. S., David, W. B. and William, L. M. (1991), The Effect of Taxes on the Relative

Valuation of Dividends and Capital Gains: Evidence from Dual-Class British

Investment Trusts, Journal of Finance 46, 383-399.

Ang, J.S.,(1987) ed. Do Dividends Matter? A Review of Corporate Dividend Theories and

Evidence (Salomon Brothers Center for the Study of Financial Institutions and the

Graduate Schools of Business Administration of New York University, New York).

Anil, K. & Kapoor, S. (2008). Determinants of Dividend Payout Ratios: A Study of

Indian Information Technology Sector. International Research Journal of

Finance & Economics 15, 64 -71.

Annand, J., Capron, L., & Will, M. (2005). Using Acquisitions to Access

Multinational Diversity: Thinking Beyond the Domestic versus Cross-Border

Comparison. Industrial and Corporate Change 14(2), 191 -224.

Appelbe, T.W.; Dineen, C.R.; Solvason, D.L. & Hsiao, C. (1992) Econometric

Modelling of Canadian Long Distance Calling: A Comparison of Aggregate

Time Series versus Point – to- Point Panel Data Approaches. Empirical

Economics 17, 125-140.

Asquith, P. and David, W. M., Jr., (1986) Signalling with, Dividends, Stock Repurchases,

and Equity Issues, Financial Management 15, 27-44.

Atrill, P. & McLaney, E. (2004). Accounting and Finance for Non –Specialists 4th

ed.

London, Pearson Education.

Auerbach, A.J. (1979). Wealth Maximisation and the Cost of Capital. Quarterly

Journal of Economics 93, 433 – 446.

Auerbach, A. J., (1983), Stockholder Tax Rates and Firm Attributes, Journal of Public

Economics 21, 107-127.

Aw, M. & Chatterjee, R. (2004). The Performance of UK Firms Acquiring Large

Cross Border & Domestic Takeover Targets. Applied Financial Economics 14,

337 -349.

184

Azhagaiah, R. & Sabari, P.N. (2008). The Impact of Dividend Policy on Shareholders‟

Wealth. International Research Journal of Finance& Economics, 180 -187.

Bajaj, M. B., and Anand, M. V. (1990), Dividend Clienteles and the Information

Content of Dividend Changes, Journal of Financial Economics 26, 193-219.

Baker, H. K., Gail, E. F. and Richard, B. E. (1985), A Survey of Management Views on

Dividend Policy, Financial Management 14, 78-84.

Baker, H. K., and Gary, E. P. (1999), How Corporate Managers View Dividend Policy,

Quarterly Journal of Business and Economics 38, 17-35.

Baker, H. Kent, G., E. P. and Theodore, V., (2002b), Revisiting the Dividend Puzzle: Do All of

the Pieces now fit?, Review of Financial Economics 11, 241-261.

Baker, H.K. & Powell, G.E. (2000a). Revisiting Managerial Perspectives on Dividend

Policy. Journal of Economics 73(2), 271- 288.

Baker, H.K. & Powell, G.E. (2000b). Determinants of Corporate Dividend Policy: A

Survey of NYSE Firms. Financial Practice and Education 9, 29 -40.

Baker, H. K., Powell, G.E. & Veil, T.E. (2001). Factors Influencing Dividend Policy

Decisions of NASDAQ Firms. The Financial Review 19-38.

Baker, J. B. (2002), Mavericks, Mergers, and Exclusion: Proving Coordinated

Competitive Effects under the Antitrust Laws. New York University Law Review

77(1), 135-203.

Baker, M. & Wurgler, J. (2004) A Catering Theory of Dividends. The Journal of Finance 59

(3) 1125-1160.,

Bali, R. (2003). An Empirical Analysis of Stock Returns Around Dividend Changes.

Applied Economics 35, 51-61.

Ball, R., Philip, B., Frank, F.J. & Oficer, R.R. (1979) Dividend and the Value of the

Firm: Evidence from the Australian Equity Market. Australian Journal of

Management 4, 13 – 26.

Baltagi, B.H. & Raj, B. (1992) A Survey of Recent Theoretical Developments in the

Econometrics of Panel Data. Empirical Economics 17, 85 -109.

Baltagi, B.H. (2001) Econometric Analysis of Panel Data. Sussex; John Wiley &

Sons Ltd.

Baradwaj, B.G., Dubofsky, D.A.and Fraser, D.R. (1992) Bidder Returns in Interstate

& Intrastate Bank Acquisitions. Journal of Financial Services Research, 261-274.

185

Barber, B. M. & Lyon, J. D. (1997). Detecting long-run abnormal stock returns: The

empirical power and specification of test statistics. Journal of Financial

Economics, 43, 341 -372.

Barclay, M. J., Clifford W. S, and Ross L. W. (1995). The Determinants of Corporate

Leverage and Dividend Policies, Journal of Applied Corporate Finance 7, 4-19.

Barclay, M. J. (1987), Dividends, Taxes, and Common Stock Prices: The Ex-Dividend

Day Behavior of Common Stock Prices Before the Income Tax, Journal of Financial

Economics 19, 31-44.

Bartholdy, J., and Kate, B. (1999), Ex-Dividend Day Pricing in New Zealand, Accounting

and Finance 39, 111-129.

.

Baskin, J. B., (1988), The Development of Corporate Financial Markets in Britain and the

United States, 1600-1914: Overcoming Asymmetric Information, The Business

History Review 62, 199-237.

Baskin, J. B., and Miranti, P. (1997). A History of Corporate Finance Cambridge,

Cambridge University Press.

Baumol, W. J., (1963), On Dividend Policy and Market Imperfection, Journal of

Business 36, 112-115.

Baumol, W.J. & Ordover, J.A. (1985), Use of Antitrust to subvert Competition.

Journal of Law & Economics 28(2), 247-265.

Baskin, J. B. (1988) The Development of Corporate Financial Markets in Britain and

the United States, 1600-1914: Overcoming Asymmetric Information Business

History Review 62, 199-237.

Beaver, W., Eger, C., Ryan, S. & Wolfson, M. (1989). Financial Reporting,

Supplemental Disclosures and Bank Share Prices. Journal of Accounting

Research 27(2), 154 – 162.

Becker, D. A. (2000). The Valuation Effects of Bank Mergers. Journal of Corporate

Finance 6,189 -214.

Beitel , P. & Schiereck, D. (2004) Value creation of domestic & cross border M&A

in the European banking market. ICFAI Journal of Mergers & Acquisitions, 3,

7-29.

Benartzi, Shlomo, Roni Michaely, and Richard H. Thaler, 1997, Do Changes in Dividends

Signal the Future or the Pst? Journal of Finance 52, 1007 – 1034.

Bernheim, B. D., (1991), Tax Policy and the Dividend Puzzle, Rand Journal of Economics

22, 455-476.

186

Bernstein, P.L., (1996), Dividends: The Puzzle, Journal of Applied Corporate Finance 9,

16-22.

Benito, A. & Young, G. (2003). Hard Times or Great Expectations? Dividend

Omissions and Dividend Cuts by UK Firms. Bank of England Working Papers

November.

Berger, A.N. & Humphrey, D.B. (1992) Measurement and Efficiency Issues in

Commercial Banking (in) Z. Griliches, ed., Output Measurement in the Service

Sector. Chicago, University of Chicago Press 245 -279.

Berger, A.N., Demesetz, R. & Strahan, P. (1999). The Consolidation of the Financial

Services Industry: Causes, Consequences, and Implications for the Future.

Journal of Banking & Finance 23, 135 -194.

Berger, A. DeYoung, R., Genay, H. & Udell, G.F. (2000). Globalisation of Financial

Institutions: Evidence from Cross Border Banking Performance. The Brookings-

Wharton Papers on Financial Services 3, 123 -158.

Berger, A. (2000) The intergeration of the financial services: where are the

efficiencies? The Federal Reserve Board of Governors Finance & Economics

Discussion Paper 2000 -36.

Berger, A. N., DeYoung, R., Genay, H. & Udell, G.F. (2001). Efficiency Barriers to

the Consolidation of the European Financial Services Industry. European

Financial Management 7, 117 -130.

Bergh, D.D. (2001) Executive Retention and Acquisition Outcomes: A Test of

Opposing Views on the Influence Organisational Tenure. Journal of

Management 27, 603-622.

Bertrand, O. & Zitouna, H. (2008). Domestic Versus Cross Border Acquisitions:

Which Impact on the Target Firm‟s Performance? Applied Economics 40 (17),

2221-2238.

Bhattacharya, S. (1979), Imperfect Information, Dividend Policy, and "the Bird in the

Hand" Fallacy, Bell Journal of Economics 10, 259-270.

Bishop, S. R., Harvey, R. C., Robert, W. F., and Garry, J. T. (2000). Corporate

Finance, Sydney, Prentice Hall.

Black, Fischer, (1976), The Dividend Puzzle, Journal of Portfolio Management 2, 5-8.

Black, F. and Scholes, M. S. (1974), The Effects of Dividend Yield and Dividend Policy

on Common Stock Prices and Returns, Journal of Financial Economics 1, 1-22.

Black, F. and Scholes, M.S. (1974). The Effects of Dividend Yield and Dividend

Policy on Common Stock Prices and Returns, Journal of Financial Economics

1, 1-22.

187

Black, E.L., Carnes, T.A., Jandik, T. & Henderson, B.C. (2003). International

Accounting Diversity and Long Term Success of Cross Border Mergers.

Working paper, University of Arkansas.

Block, S. B. (1999) A Study of Financial Analysts: Practice and Theory. Association

for Investment Management & Research.

Blume, M.E. (1980) Stock Returns and Dividend Yields: Some More Evidence.

Review of Economics and Statistics 62, 567 - 577.

Boeder, T.L. & Dorman, G.J. (2000). The Boeing/McDonnell Douglas Merger: the

Economics, Antitrust Law and Politics of the Aerospace industry. Antitrust

Bulletin 45(1) 119-152.

Bolster, P. and Janjigian, V. (1991) Dividend policy and valuation effects of the Tax

Reform Act of 1986. National Tax Journal 44 (1991), pp. 511–518.

Bonin J.P. and I. Abel, I. (2000): "Retail Banking in Hungary: A Foreign Affair?"

William Davidson Institute Working Paper 356.

Booth, L. D., and Johnston, D. J. (1984), The Ex-Dividend Day Behavior of Canadian

Stock Prices: Tax Changes and Clientele Effects, Journal of Finance 39, 457-476.

Borokhovich, K., Parrino, R. and Trapani, T. (1996), Outside Directors and CEO

Selection. Journal of Financial and Quantitative Analysis 31, 337-355.

Boyd, H. J. & Graham, S. L. (1991).Risks, Regulations and Bank Holding Company

Expansion into Non-banking. Federal Reserve Bank of Minneapolis.

Quarterly Review 10, 2-17.

Bradley, M., Capozza, D. R & Seguin, J. P. (1998) Dividend Policy and Cash-Flow

Uncertainty, Real Estate Economics 26, 555-580.

Brealey, R.A., Myers, S.C. & Marcus, A.J.(1999). Fundamentals of Corporate

Finance 4th

ed. London; Prentice Hall.

Brealey, N. (1997). Managing without management

Brennan, M. J. (1970) Taxes, Market Valuation and Corporate Financial Policy,

National Tax Journal 23, 417-427.

Bresman, H., Birkinshaw, J., & Nobel, R. (1999). Knowledge transfer in international

acquisitions. Journal of International Business Studies, 30(3), 439–462.

Brewer, E., Jackson, W., Jagtiani, J. and Nguyen, T. (2000). The price of bank

mergers in the 1990s, Federal Reserve Board of Chicago Economic

Perspectives, Q1,2–23.

188

Brigham, E. F., and Gordon, M.J. (1968), Leverage, Dividend Policy, and the Cost of

Capital, Journal of Finance 23, 85-103.

Brook, Y., Charlton, W.T. & Hendershott, R. (1998) Do Firms Use Dividends to

Signal Large Future Cash Flow Increases? Financial Management 27 93), 46 -57.

Brown, S. & Warner, J. (1985) Using Daily Stock Returns: The Case of Event Studies.

Journal of Financial Economies 14, 3-31.

Brown, P. and Walter, T. (1986), Ex-Dividend Day Behaviour of Australian Share Prices,

Australian Journal of Management 11, 139-152.

Brunarski, K., Harman, Y. & Kehr, J.B.(2004) Agency costs and the dividend

decision. Corporate Ownership & Control 1(3), 44-60

Bruner, R. F. (2005). Profiles of Outlying M&A Transactions. Working Paper 1985 –

2000.

Brutton, G .D. Oviatt, B.M. & White, M.A. (1994). Performance of Acquisitions of

Distressed Firms. Academy of Management Journal 37, 972 -989.

Buch, C. M. & Delong, G. (2004). Cross Border Bank Mergers: What Lures the Rare

Animal? Journal of Banking & Finance 28, 2077 -2102.

Byowitz, .M.C., Gottis, I.K., Katz, D.A., & Sherman, L.S. (2002) EU Proposes

Comprehensive Changes to Merger Regulations. Venulex Legal Summaries,

December 20.

Byrd, J. & Hickman, K. (1992) Do outside Directors Monitor Managers? Evidence

from Tender Offer Bids. Journal of Financial Economics 32, 195 -222.

Cabral, I., Dierick, F. & Vesala (2002). Banking Integration in the Euro Area. ECB

Occasional Paper No. 6.

Cadbury Committee (1992), Report on the Financial Aspects of Corporate

Governance, Gee,London.

Campa, J.M. & Hernand, I. (2006) M&As performance in the European financial

industry. Journal of Banking & Finance 30,3367 -3392.

Capron, L. & Pistre, N. (2002). When do Acquirers Earn Abnormal Returns? Strategic

Management Journal 22, 781 -794.

Capron, L. (1999). The Long-term Performance of Horizontal Acquisitions. Strategic

Management Journal 20 (11), 987 -1018.

Carow, K., Heron, R. & Saxton, T. (2004). Do Early Birds get the Returns? An

189

Empirical Investigation of Early Mover Advantages in Acquisitions.

Strategic Management Journal 25, 563 -577.

Casey, M.K. & Dicken, R.N. (2000) The effects of tax and regulatory changes on

commercial banks dividend policy. The Quarterly Review of Economics &

Finance 40, 279 -293.

Carruthers, Q. (2007). Banking World Rocks. Acquisition Monthly 278, 54, 55.

Certo, S., Dalton, C., Dalton, D. & Lester, R. (2008). Boards of Directors‟ Self

Interest: Expanding for Pay in Corporate Acquisitions? Journal of Business Ethics

77(2), 219-230.

Chander, S. & Priyanka, A. (2007) Determinants of Corporate Growth: An Empirical

Study of Indian Drugs and Pharmaceutical Industry. The ICFAIAN Journal of

Management Research 6 (10), 50-70.

Chang, R.P. & Rhee,G. (1990) The impact of personal taxes on corporate dividend

policy and capital structure decisions. Financial Management 19,21 -31.

Chen, Z., Heung, Y. Stouraitis, A. & Wong, W.S.W. (2005). Ownership

Concentration, Firm Performance and Dividend Policy in Hong Kong.

Pacific-Basin Finance Journal 13, 431-449.

Chen, N. Grundy, B. & Stambaugh, R.F.(1990). Changing Risk, Changing Risk

Premiums and Dividend Yield Effects. Journal of Business 63(1), 851-892.

Choi, S., I. Hasan, and B. Francis (2007). Cross-Border M&As and Bank Stability:

Evidence from the Bond Market. Rensselaer Polytechnic Institute, New York

Working paper.

Chong, B., Liu, M. & Tan, K. (2006). The Wealth Effect of Forced Bank Mergers and

Cronyism. Journal of Banking & Finance 30, 3215 – 3233.

Christiansen, A. (2006). The Reform of EU Merger Control: Fundamental Reversal or

Mere Refinement? University of Marburg working paper, April.

Christiansen, A. (2005). The "more economic approach" in EU merger control: A

critical Assessment. Deutsche Bank Research Working Paper 21.

Christiansen, A. & Kerber, W. (2005), Competition Policy with Optimally

Differentiated Rules Instead of "Per se Rules vs. Rule of Reason", available at

http://ssrn.com/abstract=872694(24/05/2008)

Coakley, J. & Iliopoulou, S.( 2006). Bidder CEO and Other Executive Compensation

in UK M&As. European Financial Management 12(4), 609-631.

190

Coate, M. B. & Kleit, A.N. (1998). Does it Matter that the Prosecutor is Also the

Judge? The Administrative Complaint Process at the Federal Trade Commission.

Managerial & Decision Economics 19(1), 1-11.

Colley, L. (2004), From „Defense‟ to „Attack‟? Quantifying Efficiency Arguments in

Mergers. European Competition Law Review, 25(6), 342-349.

Conn, R.L. (2003). International Acquisitions: Review of Literature and Clinical

Projects. Journal of Financial Education 29, 1 -27.

Conn, R.L., Cosh, A. Guest, P.M. & Hughes, A. (2005). The Impact of UK Acquirers

of Domestic, Cross Border, Public and Private Acquisitions. Journal of Business

Finance & Accounting 32 (5&6), 815 -869.

Conroy, R. M., Kenneth, M. E. and Robert, S. H., (2000), A Test of the Relative Pricing

Effects of Dividends and Earnings: Evidence from Simultaneous Announcements in

Japan, Journal of Finance 55, 1199-1227.

Cornett, M.M. & Tehradnian, H. (1992) Changes in Corporate Performance Associated with

Bank Acquisitions. Journal of Financial Economics 31, 211-234.

Cotter, J. Shivdasani, A., Zenner, M. (1997) Do Independent Directors Enhance Target

Shareholder Wealth During Tender Offers? Journal of Financial Economics 43,

195 -218.

Cross, M. L., Davidson, W. N. & Thornton, J.H. (1988). Taxes, Stock Return and

Captive Insurance Subsidiaries. Journal of Risk and Insurance 55, 331 -338.

Cybo-Ottone, A. and Murgia, M. (2000) Mergers & shareholders wealth in in

European banking. Journal of Banking and Finance, 24 ,831 -859.

Dammers, C. (2008). Banks go back to the Future. Euroweek, 13, 14.

Danbolt, J. (1995). An Analysis of Gains and Losses to Shareholders of Foreign

Bidding Companies Engaged in Cross Border Acquisitions into the United

Kingdom. European Journal of Finance 1, 279 -309.

Davidson, W. N. & Cheng, L.T.W. (1997) Target Firm Returns: Does The Form of

Payment Affect Abnormal Returns? Journal of Business Finance & Accounting,

24(3& 4), 465 -481.

Davies, T. & Pain, B.(2002). Business Accounting & Finance. London,

McGraw-Hill.

Delong, G. & DeYoung, R. (2007) Learning by Observing: Information Spillovers in

the Execution and Valuation of Commercial Banks M&As. The Journal of

Finance LXII (1), 181-216.

191

DeAngelo, H., and R.W. Masulis(1980), "Optimal Capital Structure under Corporate

and Personal Taxation," Journal of Financial Economics 8, 3-29.

DeAngelo, H.,DeAngelo, L. & Stilz, R.M. (2006) Dividend policy & the earned/

contributed capital mix: a test of the life cycle theory. Journal of Financial

Economics 81 ,227-254.

DeAngelo, H., DeAngelo, L. and Skinner, D.J. (1992), Dividends and Losses, Journal of

Finance 47, 1837-1863.

Dempsey, S. J., and Laber, G. (1992), Effects of Agency and Transaction Costs on

Dividend Payout Ratios: Further Evidence of the Agency-Transaction Cost

Hypothesis, Journal of Financial Research 15, 317-321.

Dempsey, S. & Laber, G. (1992) Effect of Agency and Transactions Costs on

Dividend Payout Ratios: Further Evidence of the Agency –Transition Cost

Hypothesis. Journal of Financial Research 15, 317 -321.

Demsetz, H. & Lehn, K. (1985). The Structure of Corporate Ownership: Causes and

Consequences. Journal of Political Economy 93(6), 1155 -77.

Denis, D. J., Denis, K.D. and Sarin, A. (1994), The Information Content of Dividend

Changes: Cash Flow Signaling, Overinvestment, and Dividend Clienteles, Journal of

Financial and Quantitative Analysis 29, 567-587.

DePamphilis, D. (2003). Mergers, Acquisitions, and Other Restructuring Activities:

An Integrated Approach to Process, Tools, Cases, and Solutions, 2nd

ed., San

Diego, Elsevier/Academic Press.

Dewenter, K. L., and Warther, V. A. (1998), Dividends, Asymmetric Information, and

Agency Conflicts: Evidence from a Comparison of the Dividend Policies of Japanese

and U.S. Firms, Journal of Finance 53, 879-904.

DeYoung, R. (1997). Bank mergers, X-efficiency, and the market for corporate

Control. Managerial Finance 23, 32-47.

DeYoung, R. (1999). Birth, growth, and life or death of newly chartered banks.

Federal Reserve Bank of Chicago. Economic Perspectives 23 (3), 18-35.

Dhaliwal, D. S., Erickson, M. and Trezevant, R.(1999), A Test of the Theory of Tax

Clienteles for Dividend Policies, National Tax Journal 52, 179-194.

Diamond, J. J. (1967), Earnings Distribution and the Valuation of Shares: Some Recent

Evidence, Journal of Financial and Quantitative Analysis 2, 15-30.

Diaz, B., Olla, M.G. & Azofra, S.S. (2004) Bank Acquisitions and Performance:

Evidence from a Panel of European Credit Entities. Journal of Economics &

Business 56, 377 -404.

192

Dieleman, F.M. (2001). Panel Data in Housing Research: A Reaction to the Paper by

Myers on Cohort Longitudinal Estimation of Housing Careers. Housing Studies

16(1), 115 -121.

Dimson, E., Marsh, P. and Staunton, M. (2002) Triumph of the Optimists: 101 Years

of Global Investment Returns, Princeton University Press.

Doughty, M. (2000). The Joy of Money. London; Simon & Schuster UK Ltd.

Drauz, G. (2002). A View from Inside the Merger Task Force: Comments on

„Reforming European Merger Review: Targeting Problem Areas in Policy

Outcomes‟. Journal of Industry, Competition and Trade, 2(4), 391-399.

Easterbrook, F.H. (1984). Two Agency Cost Explanations of Dividends. American

Economic Review 74, 650 -659.

Easton, P. (2004), „PE Ratios, PEG Ratios, and Estimating the Implied Expected Rate

of Return on Equity Capital‟. The Accounting Review 79, 73–96.

Eckbo, B.E. (1983). Horizontal Mergers, Collusion and Stockholder Wealth. Journal

of Financial Economics 11(1&4), 241 -273.

Ecko, E.B. & Thornburn, K.S. (2000). Gains to Bidder Firms Revisited: Domestic and

Foreign Acquisitions in Canada. Journal of Financial and Quantitative Analysis

35, 1-25.

Economic Commission Competition Policy Annual Report (2007)

Elton, E. J., and Gruber, M. J. (1970), Marginal Stockholder Tax Rates and the Clientele

Effect, Review of Economics and Statistics 52, 68-74.

Elton, E. J., Gruber, M. J.and Blake, C.R. (2002), Marginal Stockholder Tax

Effects and Ex-Dividend Day Behavior: Thirty-Two Years Later, New York

University, Stern School of Business Working Paper No. 02-047.

Elton, E. J., Gruber, M. J. and Rentzler, J. (1984), The Ex-Dividend Day Behavior of

Stock Prices; A Re-Examination of the Clientele Effect: A Comment, Journal of

Finance 39, 551-556.

European Economic Research 1996

Faccio, M., Lang, L.H.P.,& Young, L.(2001). Dividends and Expropriation. American

Economic Review 91, 54-78.

Fama, E.F. & French, K.R. (2001). Disappearing Dividends: Changing Firm

Characteristics or Lower Propensity to Pay? Journal of Financial Economics 60,

3- 43.

Fama E.F.& French, K.R.(2003). The Capital Asset Pricing Model: Theory and

Evidence. Social Science Research Network Electronic Library, CRSP

Working Paper, No. 550.

193

Fama, E. F. (1980). Agency problems and the theory of the firm. Journal of Political

Economy 88 (2), 288-307.

Fama, E. F., and Babiak, H. (1968), Dividend Policy: An Empirical Analysis, Journal of

American Statistical Association 63, 1132-1161.

Fazzari, S. M., Hubbard, R.G. and Petersen, B.C. (1988), Financing Constraints and

Corporate Investment, Brooking Papers on Economic Activity 1, 141-195.

Farrar, D. E. and Lee L. Selwyn, L.L.(1967). Taxes, Corporate Financial Policy and

Return to Investors, National Tax Journal, 20, 444-462.

Feldstein, M. and Green, J. (1983) Why do companies pay dividend? The American

Economic Review 73(1), 17 -30.

Finerran, J. (2006). Bank Efficiency: Measure with Care. (Online)

http://www.fool.com/investing/dividends-income/2006/12/14/bank-efficiency-measure-with-

care.aspx (14/12/2006).

Forcarelli, D. & Pozzolo, A.F. (2001). The Patterns of Cross Border Bank Mergers

& Shareholdings in OECD Countries. Journal of Banking and Finance 25,

2305 – 2337.

Frankfurter, G.M. & Lane, W.R. (1992). The Rationality of Dividends. International

Review of Financial Analysis 1, 115 -129.

Frankfurter, G. M., Arman, K, Hartmut, S., and Mihail, T. (2002), The Perception of

Dividends By Management, Journal of Psychology and Financial Markets 3, 202-217.

Frankfurter, G. M., and Wood, B.G. Jr.,( 1997), The Evolution of Corporate Dividend

Policy, Journal of Financial Education 23, 16-33.

Friend, I. and Puckett, M. (1964), Dividend and Stock Prices, American Economic

Review 54, 656-682.

Gaughan, P.A. (2002) Mergers, Acquisitions and Corporate Restructuring.

New York; Wiley.

Gaver, J.J. & Gaver, K.M. (1993). Additional Evidence on the Association between

the Investment Opportunity Set and Corporate Financing, Dividend, and

Compensation Policies. Journal of Accounting and Economics 16, 125 -160.

Gilson, S. (1990). Bankruptcy, Boards, Banks, and Blockholders: Evidence on

Changes in Corporate Ownership and Control when Firms Default. Journal of

Financial Economics 26, 355- 387.

Goddard, J., Molyneux, P. & Wilson, J.O.S. (2004a). The Profitability of European

Banks: A Cross –sectional and Dynamic Panel Analysis. The Manchester

School, 72 (3), 363 -381.

194

Goddard, J., Molyneux, P. & Wilson, J.O.S. (2004b) Dynamics of Growth and

Profitability in Banking. Journal of Money, Credit and Banking, 36(6),

1069 -1090.

Goddard, J., Molyneux, P., Wilson, J.O.S. & Tavakoli (2007) European banking: An

Overview. Journal of Banking and Finance 31, 1911 -1935.

Gordon, M.J. (1959). Dividends, Earnings, and Stock Prices. The Review of

Economics and Statistics 41, 99 -105.

Gordon, M. J., (1963), Optimal Investment and Financing Policy, Journal of Finance 18,

264-272.

Gordon, M. J., and Shapiro,E. (1956), Capital Equipment Analysis: The Required Rate of

Profit, Management Science 3, 102-110.

Grammatikos, T. (1989), Dividend Stripping, Risk Exposure, and the Effect of the 1984

Tax Reform Act on the Ex-Dividend Day Behavior, Journal of Business 62, 157-173.

Gregory, A., & McCorriston, S. (2004). Foreign acquisitions by UK limited

companies: Short and long-run performance. Working paper, University of Exeter.

Grossman, S. J. and Hart, O. D, Moore, D. (1986). The Costs and Benefits of

Ownership: A Theory of Vertical and Lateral Integration," Journal of Political

Economy, 94 (4), 691-719.

Grullon, G., Michael, R. and Swaminathan, B. (2002) Are Dividend Changes a Sign of

Firm Maturity? Journal of Business 75, 387-424.

Gulger, K. & Yurtoglu, B.B. (2003) Corporate governance & dividend payout policy

in Germany. European Economic Review 47,731-758.

Gugler, K. (2004). Corporate Governance, Dividend Payout Policy, and the

Interrelation between Dividends, R&D, and Capital Investment. Journal of

Banking & Finance 27, 1297 – 1321.

Hage, D. (1991). Frightened Financiers. US News & World Report 111(5), 40.

Hair, J.F.; Black, W.C.; Barry, J.B.; Anderson, R.E. & Tatham, R.L. (2005).

Multivariate Data Analysis 6th

ed., New Jersey, Pearson Education.

Han, K. C., Lee, S. H. & Suk, D.y. (1999). Institutional Sahreholders and Dividends.

Journal of Financial & Strategic Decision 12, 53 -62.

Hany, A. (2005). Dividend Policy of Quoted Firms in Emerging Markets, the case of

Jordan. Unpublished PhD thesis submitted to the Department of Economics &

Finance, University of Western Sydney.

195

Harris, T., Hubbard, G. & Kemsley, D. (1997). Are Dividend Taxes and Tax

Imputation Credits Capitalised in Share Values? Manuscript, Columbia University.

Harris, C.P. (2007) European Merger Review: Tightening Control While Expanding

Third Party Power. North Carolina Journal of Internal Law & Commercial

Regulations 33(2), 338 -357.

Harrison, J. S., Oler, D., & Allen, M.R. (2005). Event Studies and the Importance of

Longer –Term Measures in Assessing the Performance Outcomes.Unpublished

Manuscript.

Hawawini, G. and Swary, I. (1990) Mergers and Acquisitions in the U.S. Banking

Industry; Evidence from the Capital Markets. North-Holland, Amsterdam.

Hayward, M. L. A. (2002). When do firms learn from their acquisition experience?

Evidence from 1990–1995. Strategic Management Journal, 23(1), 21–39.

Heath, D. C., and Jarrow, R.A. (1988), Ex-Dividend Stock Price Behavior and

Arbitrage Opportunities, Journal of Business 61, 95-108.

Henderson, G. V. (1989) Problems and Solutions in Conducting Event Studies.

Conference Paper presented at the American Risk and Insurance Association,

Denver, August.

Hermalin, B.E. & Weisbach, M.S. (1991). The Effects of Board Composition and

Direct Incentives on Firm Performance. Financial Management, 101 -112.

Hernando, I., Nieto, M.J. & Wall, L.D. (2008).Determinants of Domestic and Cross

Border Bank Acquisitions in the European Union. Working paper series 2008 -26,

Federal Reserve Bank of Atlanta.

Hess, P. J.(1981) ed.The Dividend Debate: 20 Years of Discussion, In "The Revolution in

Corporate Finance", 1992 (Blackwell Publishers, Cambridge, Massachusetts).

Hietala, P. T., (1990), Equity Markets and Personal Taxation: the Ex-Dividend Day

Behaviour of Finnish Stock Prices, Journal of Banking and Finance 14, 327-350.

Hitt, M., Hoskisson, R.E., Ireland, R.D., & Harrison, J.S. (1991). Effects Acquisitions

on R&D Inputs and Outputs. Academy of Management Journal 34 (3), 693 -706.

Hitt, M., Harrison, J., Ireland, R.D., & Best, A. (1998). Attributes of Successful and

Unsuccessful Acquisitions US Firms. British Journal of Management

9, 91 - 114.

Hitt, M., Hoskisson, R. E., Johnson, R. & Moesel, D. (1996). The Market for

Corporate Control & Firm Innovation. Academy of Management Journal 39,

1084 -1109.

Ho, H. (2003) Dividend Policies in Australia and Japan, International Advances in

Economic Research 9, 91-100.

196

Houston, J. & Ryngaert, M. (1994) The Over Gains from Large Bank Mergers.

Journal of Banking & Finance 18, 1155-1176.

Huang, Y. and Walking R. (1987): “Target abnormal returns associated with

acquisition announcements payment, acquisition form and managerial

resistance”, Journal of Financial Economics, 19, p 329-349.

Hughes, J., Lang, W., Mester, L., Moon, C.G (1999). The Dollars and Sense of Bank

Consolidation. Journal of Banking & Finance 23, 291 -324.

Hughes, J.P., Lang, W.W., Mester, L.J., Moon, C.-G. and Pagano, M.S. (2003). Do

bankers Sacrifice value to build empires? Managerial incentives, industry

consolidation, and Financial performance. Journal of Banking and Finance

27 (3), 417–447.

Hutchinson, W. R. (1995). The capital structure and investment decisions of the

Small owner-managed firm: Some exploratory issues. Small Business

Economics 7(3), 231-239.

James, C. & Huston, J. (1995) CEO Compensation and Bank Risk: Is Compensation in

Banking Structured to Promote Risk Taking? Journal of Monetary Economics

36,405 -431.

Jankowicz, A. D. (2000) Business Research Projects Thomson, Business Press.

Jensen, M.C. (1986). Agency Costs of Free Cash Flow, Corporate Finance, and

Takeovers. American Economic Review 76, 323 -29.

Jensen, G. R.,Solberg, D. S. and Zorn, T. S. (1992), Simultaneous Determination of

Insider Ownership, Debt, and Dividend Policies, Journal of Financial and

Quantitative Analysis 27, 274-263.

Jensen, M. C., (1986), Agency Costs of Free Cash Flow, Corporate Finance, and

Takeovers, American Economic Review 76, 323-329.

Jensen, M. C., and Meckling, W. H. (1976), Theory of the Firm: Managerial Behavior,

Agency Costs and Ownership Structure, Journal of Financial Economics 3, 305-360.

John, K., and Williams, J. (1985), Dividends, Dillution, and Taxes: A Signalling

Equilibrim, Journal of Finance 40, 1053-1070.

Johnson, S.A., Lin, J.& Song, K.R. (2006). Dividend Policy, Signalling, and Discounts

on Closed End Funds. Journal of Financial Economics 81, 539 -562

197

198

Kalay, A. and Michaely, R. (2000), Dividends and Taxes: A Re-examination, Financial

Management 29, 55 -75.

Kalay, A. (1980). Signalling, Information Content, and the Reluctant to Cut

Dividends. Journal of Financial and Quantitative Analysis 15,855-869.

Kanatas, G. & Qi, J. (2004). Dividends and Debt with Managerial Agency and Lender

Hold-up. Management Science, 1249 -1260.

Kane, E. (2000) Incentives for banking megamergers: what motives might

regulators infer from event-study evidence? Journal of Money, Credit &

Banking 32.

Kang, H. & Diltz, D. (2000) Dividend announcements and the valuation effects of

corporate divestiture. Journal of Financial & Strategic Decision, 13(2), 25-32.

Kapoor, R. & Lim, K. (2005). The Impact of Acquisition on the Innovation

Performance of Inventors at Semiconductor Companies. Academy of

Management Journal 50(5), 1133- 1155.

Karpoff, J. M., and Walkling, R.A. (1988), Short-Term Trading Around Ex-Dividend

Days: Additional Evidence, Journal of Financial Economics 21, 291-298.

Karpoff, J. M., and Walkling, R.A. (1990), Dividend Capture in NASDAQ Stocks,

Journal of Financial Economics 28, 39-65.

Kato, K., and Loewenstein, U. (1995), The Ex-Dividend-Day Behavior of Stock Prices:

The Case of Japan, The Review of Financial Studies 8, 817-847.

Kaufmaann, Gordon & Owers 2000

Keeton, W. (2000) Are mergers responsible for the surge in new bank charters.

Economic Review, 1st quarter, 21-41.

Keim, D. B., (1985), Dividend Yields and Stock Returns: Implications of Abnormal

January Returns, Journal of Financial Economics 14, 473-489.

Koch, P. D., and Shenoy, C. (1999), The Information Content of Dividend and Capital

Structure Policies, Financial Management 28, 16-35.

Kolb, R.W. & Rodriguez, R.J. (1996). Financial Markets. Oxford: Blackwell Business

Koski, J. L., and Scruggs, J. S. (1998), Who Trades Around the Ex-Dividend Day?

Evidence from NYSE Audit File Data, Financial Management 27, 58-72.

Kumar, R. and Sopariwala,P. R. (1992). The Effect of Adoption of Long-Term

Performance Plans on Stock Price and Accounting Numbers, Journal of

Financial and Quantitative Analysis 27, 561-573.

199

Kwan, S. H. & Wilcox, J.A. (1999). Hidden Cost Reductions in Bank Mergers:

Accounting for more Productive Bank. Federal Reserve Bank of Chicago, May,

533-547.

Larsson, R. & Finkelstein, S. (1999). Integrating Strategic, Organisational, and Human

Resources Perspective on Mergers and Acquisitions: A Case Survey Synergy

Realization. Organisation Science 10(1), 1 -26.

La Porta, R., Florenico, L., Shleifer, A. and Vishny, R. W. (2000a), Agency Problems and

Dividend Policies Around the World, Journal of Finance 55, 1-33.

Lakonishok, J., and Vermaelen, T. (1983), Tax Reform and Ex-Dividend Day Behavior,

Journal of Finance 38, 1157-1179.

Lease, R. C., Kose J., Avner, K., Uri, L., and Oded, H. S. (2000). Dividend Policy:

It’s Impact on Firm Value. Massachusetts, Harvard Business School Press.

Lensink, R. & Maslennikova, I. (2008). Value Performance of European Bank

Acquisitions. Applied Financial Economics 18, 185 -198.

Levy, N. (2003) EU Merger Control: From Birth to Adolescence. World Competition

26 (2), 195 -218.

Lewellen, W. G., Stanley, K.L., Lease, R.C. and Schlarbaum, G.G. (1978), Some Direct

Evidence on the Dividend Clientele Phenomenon, Journal of Finance 33,1385-1399.

Lie, E. (2000), Excess Funds and Agency Problems: An Empirical Study of Incremental

Cash Disbursements, Review of Financial Studies 13, 219-248.

Lingos, T., Loughran, M., Pitkanen, T. & Todino, M. (2004). An amended merger

Implementing Regulation for a new Merger Regime. EC Competition Policy

Newsletter 2, 79-84.

Lintner, J. (1956). Distribution of Incomes of Corporations among Dividends,

Retained Earnings and Taxes. American Economics Review 46, 97 -113.

Lintner, J. (1964) Dividends, Earnings, Leverage, Stock Prices and Supply of Capital

to Corporations. The Review of Economics and Statistics 64, 243-269.

Lipson, M. L, Maquieira, C. P., and William, M. (1994). Dividend Initiations and

Earnings Surprises. Financial Management 27, 36-45.

Lipson, M. L, Maquieira, C. P., and Megginson, W. (1998), Dividend Initiations and

Earnings Surprises, Financial Management 27, 36-45.

Litzenberger, R. H., and Ramaswamy,K. (1979), The Effect of Personal Taxes and

Dividends on Capital Asset Prices, Journal of Financial Economics 7, 163-195.

Lloyd, W. P., Jahera, S.J. and Page, D.E. (1985), Agency Costs and Dividend Payout

Ratios, Quarterly Journal of Business and Economics 24, 19-29.

200

Lowsinski, F., Schiereck, D. & Thomas, T.W. (2004). The Effect of Cross Border

Acquisitions on Shareholders Wealth – Evidence from Switzerland. Review of

Quantitative Finance & Accounting 22, 315 -330.

MacKinlay, A.C. (1997) Event studies in Economics and Finance. Journal of

Economic Literature, 35, 13-39.

Mace, M. L. (1986). Directors, Myth and Reality. Boston, Harvard Business School

Press.

Madura, J. & K.J. Wiant, K. J. (1994). Long-term valuation effects of bank

acquisitions. Journal of Banking and Finance 18 (6), 1135–1154.

Markides, C. (992) Consequences of corporate refocusing: ex ante evidence. The

Academy of Management Journal 35, 398 – 412.

Mangold, N.R. &Lippok, K. (2008). The Effect of Cross Border Mergers and

Acquisitions on Shareholders Wealth: Evidence from Germany. Journal of

International Business & Economics 8, 30 -54.

Massa, M. & Zhang, L. (2009). Stock, Bonds and Debts Imbalance: The Role of

Relative Availability of Bond and Bank Financing. INSEAD Working Paper

Series.

Masulis, R.W. and Trueman, B.(1988).Corporate investment and dividend decisions

Under differential personal taxation. Journal of Financial and Quantitative

Analysis 23,369–386.

Maury, B. and A. Pajuste (2002), “Controlling Shareholders, Agency Problems and

Dividend Policy in Finland”, Finnish Journal of Business Economics, 51, 15-45.

McManus, I., Gwilym, O.I. & Thomas, S. (2004). The Role of Payout Ratio in the

Relationship between Stock Returns and Dividend Yield. Journal of Business

Finance & Accounting 31 (9 &10), 1355 -1387.

Michaely, R., Thaler, R.H. and Womack, K. L. (1995), Price Reactions to Dividend

Initiations and Omissions: Overreaction or Drift?,Journal of Finance 50, 573-608.

Michaely, R., and Vila, J. (1996), Trading Volume with Private Valuation: Evidence from

the Ex-Dividend Day, Review of Financial Studies 9, 471-509.

Miller, M.H. (1986). Behaviour Rationality in Finance: the Case of Dividends. Journal of

Business 59, 451-468.

Miller, M. H., (1977), Debt and Taxes, Journal of Finance 32, 21-35.

201

Miller, M. H., and Modigliani, F. (1961), Dividend Policy, Growth, and the Valuation of

Shares, Journal of Business 34, 411-433.

Miller, M. H., and Rock, K. (1985), Dividend Policy Under Asymmetric Information,

Journal of Finance 40, 1031-1051.

Miller, M. H., and Scholes, M.S. (1978), Dividends and Taxes, Journal of Financial

Economics 6, 333-264.

Miller, M. H., and Scholes, S.M. (1982), Dividend and Taxes: Some Empirical Evidence,

Journal of Political Economy 90, 1118-1141.

Modigliani, F. (1982). Debt, Dividend Policy, Taxes, Inflation and Market Valuation.

Journal Of Finance 255-273.

Moeller, S.B., Schlingemann, F.P. & Stulz, R. (2004). Firm Size and the Gains from

Acquisitions. Journal of Financial Economics 73 (2), 201-228.

Moh‟d M.A., Perry, L.G. & Rimbey, J.N. (1995). An Investigation of the Dynamic

Relationship between Agency Theory and Dividend Policy. Financial Review

30, 367 - 385.

Morck, R., Schleifer, A., & Vishny, R. (1989) Management Ownership and Market

Valuation: An Empirical Analysis. Journal of Financial Economics 20 (1&2),

293-315.

Morgan, G. and Thomas, S. (1998), Taxes, Dividend Yields and Returns in the UK

Equity Market, Journal of Banking and Finance 22, 405-423.

Motta, M. (2000), EC Merger Policy and the Airtours Case. European Competition

Law Review, 21(4), 199-207.

Motta, M., Polo, M. & Vasconcelos, H. (2007) Merger Remedies in the European

Union: An Overview. The Antitrust Bulletin 52(3), 603 -630.

Mulherin, J.H. & Boone, A.L. (2000). Comparing Acquisitions and Divestitures.

Journal of Corporate Finance 6, 117 -139.

Murray, A. (2004). A fair Referee? The European Commission and EU competition

policy. Centre for European Reform Report, London.

Myers, S.C. & Majluf, N.S. (1984) Corporate Financing and Investment Decisions

When Firms Have Information that Investors do not have. Journal of Financial

Economics 13, 187 -221.

Neven, D., L. & Roller, H. (2002), Discrepancies between Markets and Regulators:

An Analysis of the First Ten Years of EU Merger Control. Graduate Institute of

International Studies Geneva, HEI Working Paper No. 10.

202

Nnadi, M.A. & Akpomi, M.E. (2008). The Effect of Taxes of Dividend Policy of

Banks in Nigeria. International Research Journal of Finance & Economics

19, 48-55.

Nissim, D. and Ziv,A. (2001), Dividend Changes and Future Profitability, Journal of

Finance 56, 2111-2133.

Ofer, A. R. & Siegel, D.R. (1987). Corporate financial policy, information, & market

expectations: an empirical investigation of dividends. Journal of Finance 42,

889- 911.

Ofer, A. R., and Thakor, A. V. (1987), A Theory of Stock Price Responses to Alternative

Corporate Cash Disbursement Methods: Stock Repurchases and Dividends, Journal

of Finance 42, 365-394.

Okazaki, T., Sawada, M. & Yokoyama, K. (2005) Measuring the Extent and

Implications of Director Interlocking in the Prewar Japanese Banking

Industry .The Journal of Economic History (2005), 65(4), 1082-1115.

Olson, G.T. & Pagano, M.S. (2005). A New Application of Sustainable Growth: A

Multi- Dimensional Framework for Evaluating the Long Run Performance of

Bank Mergers. Journal of Business Finance & Accounting 32 (9 & 10),

1995 -2036.

Omet, G. (2004). Dividend Policy Behaviour in the Jordanian Capital Market,

International Journal of Business 9, 287-299.

Omet, G. And Abu-Ruman, G. (2003), Dividend Policy in the Jordanian Capital Market:

Empirical Findings and Survey Results, Second International Finance Conference,

March 13-15 (Hammamet, Tunisia).

Ongena, S. & Penas, M.F. (2008) Bondholders‟ Wealth Effects in Domestic and Cross

Border Bank Mergers. Journal of Financial Stability (Article in press).

Padilla, A. J. (2004). The „Efficiency Offense Doctrine‟ in European Merger Control,

in: Evans, D. S., A. J. Padilla (Eds.), Global Competition Policy – Economic

Issues and Impact, Emeryville, pp. 159-168.

Page, D.E., Jahera, J.S. & Pugh, W.N. (1996). The Effect of Takeover Defences on

the Dividend Decision. Journal of Economics and Finance 20 (3), 49 -58.

Palich, L.E., Cardinal, L.B. & Miller, C.C. (2000). Curvi-linearity in the

Diversification Performance Linkage: An Examination of over Three Decades of

Research. Strategic Management Journal 21(2), 155-174.

Pallant, J. (2006). SPSS Survival Manual 2nd

ed., Berkshire; Open University Press.

Pandey, I. (2001). The Expected Stock Returns of Malaysian Firms: A Panel Data

Analysis. IIMA Working Paper No 2001-09-01.

203

Pangarkar, N. (2004). The Impact of Market Cycle on the Performance of Singapore

Acquirers. Strategic Management Journal 21(2), 155-174.

Papaioannou, G. J. and Savarese, C. M. (1994) Corporate dividend policy response to

the Tax Reform Act of 1986. Financial Management 23, 56–63.

Partington, G. H., (1985), Dividend policy and its Relationship to Investment and

Financing Policies: Empirical Evidence, Journal of Business Finance and Accounting

12, 531-542.

Pasiouras F., Tanna S., Zopounidis C., (2005), Application of Quantitative Techniques

for the Prediction of Bank Acquisition Targets. London; World Scientific

Publishing Co.

Pasurious,F, Tanna, S. & Zopounidis, C. (1997), Regulations, Supervision and Banks'

Cost and Profit Efficiency around the World: A Stochastic Frontier Approach.

University of Bath Management Working Paper 2007.05.

Pasiouras, F., Tanna, S. & Gaganis, C. (2007). What Drives Acquisitions in EU

Banking Industry? The Role of Bank Regulations and Supervision Framework,

Bank Specific and Market Specific Factors. Coventry University, Economics

Finance & Accounting Applied Research Working Papers series 2007-3.

Pattern, K. & Twite, G. (2008). Taxes and Dividend Policy under Alternative Tax

Regimes. Journal of Corporate Finance 14, 1 -16.

Peek, J. & Rosengren, S.E. (1997). Have Borrower Concentration Limits Encouraged

Bank Consolidation? New England Economic Review, Federal Reserve Bank

of Boston, Jan, 37-47.

Penas, M. F. and H. Unal (2004).Gains in Bank Mergers: Evidence from the Bond

Market. Journal of Financial Economics 74(1): 149-179.

Pennings, J.M., Barkema, H. & Douma, S. (1994). Organisational Learning and

Diversification. Academy of Management Journal 37(3), 608-640.

Perez, D., Salas, V. & Saurina, J. (2006). Earnings and capital management in

Alternative loan Loss provision regulatory regimes. Documentos de Trabaja No

0614, Banco De Espana.

Pettit, R. R. (1977) Taxes, Transactions costs and the Clientele Effect of Dividends,

Journal of Financial Economics 5, 419-436.

Pillof, S.J. & Santomero, A.M. (1997). The Value Effect of Bank Mergers and

Acquisitions . In Amihud, Y. & Miller. G. (eds.), Bank Mergers and

Acquisitions, 59 -78. Kluwer Academic.

204

Pillof, S.J. (1996) Performance changes and shareholder wealth creation associated

with mergers of publicly traded institutions. Journal of Money, Banking &

Finance;28,294- 310.

Pons, J. & Sautter, F.T. (2004), Ensuring a sound competition environment: Rules,

Practice, Reforms and Challenges of European Competition Policy, in: Eekhoff, J.

(Ed.), Competition Policy in Europe, Berlin, 29-62.

Portes, R. & Rey, H. (1999). The Determinants of Cross Border Equity Flow.

Journal of International Economics 65(2), 269 -296.

Poterba, J. (2004). Taxation and Corporate Payout Policy. Taxation, Investment and

Saving Journal 94 (2), 171-175.

Poterba, J. M., and Summers, L.H. (1984), New Evidence That Taxes Affect the

Valuation of Dividends, Journal of Finance 39, 1397-1415.

Puranam, P., Singh, H. & Zollo, M. (2006). Organising for Innovation: Managing the

Coordination –Autonomy Dilemma in Technology Acquisitions. Academy of

Management Journal 49(2), 263 -280.

Rad, A.T. & Beek, L. V. (1999) Market valuations of European bank mergers.

European Management Journal 13(5), 532-540.

Ramaswamy, K. (1997). The Performance Impact of Strategic Similarity in Horizontal

Mergers: Evidence from the US Banking Industry. Academy of Management

Journal 40(3), 697 –715.

Reuters (2008). Financial Database

--------- (2009). Financial Database

Rhoades, S.A. (1994) A Summary of bank performance srudies in banking and an

assessment of operating performance and event study methodologies.

Federal Reserve System,167.

Rhoades, S.A. (1998) Bank Mergers and Banking Structure in the United States,

1980-1998, Board of Governors of the Federal Reserve System Staff Study 174.

Richardson, G., Sefcik, S. E.and Thompson, R. (1986), A Test of Dividend

Irrelevance Using Volume Reactions to a Change in Dividend Policy, Journal of

Financial Economics 17, 313-333.

Roll, R. (1988). The Hubris Hypothesis of Corporate Takeover. Journal of Finance

55, 197 – 216.

Rosenstein, S. & Wyatt, J. (1990) Outside Directors, Board Independence, and

Shareholders Wealth. Journal of Financial Economics 26, 175 -192.

205

Ross, R.A., Westerfield, R.W. & Jordan, B.D. (2001). Essential of Corporate Finance

2nd

ed. London, McGraw-Hill.

Rozeff, M.S. (1982). Growth, Beta and Agency Costs as Determinants of Dividend

Payout Ratios. Journal of Financial Research 5 (30, 249-259.

Samuels, J. M. & Wilkes, F. M. (1997) Management of Company Finance, London:

Chapman and Hall.

Sapienza, P. (2002). The Effect of Banking Mergers on Loan Contracts. The Journal

of Finance, 52(1), 329- 367.

Serra, A. P. (2002) Event Study tests: A brief survey. Working papers da FEP no.

117, 1-10.

Scott & Dunkelberg 2003 Scott, J.A., Dunkelberg. W. (2003). Bank mergers and small

Firm financing. Journal of Money, Credit & Banking 35, 999-1017.

Scholz, J. K., (1992), A Direct Examination of the Dividend Clientele Hypothesis, Journal

of Public Economics 49, 261-285.

Seelig, S.A.& Critchfield, T.(1999). Determinants of De-Novo Entry in Banking.

Federal Deposit Insurance Corporation working paper 99-1.

Seida, J. A., (2001), Evidence of Tax-Clientele-Related Trading Following Dividend

Increases, Journal of the American Taxation Association 23, 1-21.

Seiler, M.J. (2004). Performing Financial Studies: A Methodological Cookbook. New

Jersey; Pearson Education Ltd.

Seth, A., Song, K.P. & Pettit, R. (2002). Value Creation and Destruction in Cross

Border Acquisition: An Empirical Analysis of Foreign Acquisitions of US

Firms. Journal of International Business Studies 31 (3), 387 -405.

Shahrur, H. (2005). Industry Structure and Horizontal Takeovers: Analysis of Wealth

Effects On Rivals, Suppliers, and Corporate Customers. Journal of Financial

Economics 76(1), 61 -98.

Shanley, M. T. & Carrea, M.E. (1992). Agreement between Top Management Teams

and Expectations for Post Acquisition Performance. Strategic Management

Journal 13(4), 245 -266.

Shearer, B. (2006). Battening Down for a Credit Crunch. Mergers & Acquisitions:

The Dealer maker’s Journal 41 (8), 26 -30.

Shefrin, H.M., and Statman, M. (1984). Explaining Investor Preference for Cash

Dividends. Journal of Financial Economics, 253-282.

Sheifer, A. & Vishny, R.W. (2003) Stock Market Driven Acquisitions. Journal of

Financial Economics 70, 295 -311.

206

Shivdasani, A., (1993). Board Composition, Ownership Structure, and Hostile

Takeovers. Journal of Accounting and Economics 16, 167-198.

Short, H. (1994). Ownership, Control, Financial Structure and the Performance of

Firms, Journal of Economic Surveys 8, 203-249.

Short, H. , Zhang,H. and Keasey, K. (2002), The Link Between Dividend policy and

Institutional Ownership, Journal of Corporate Finance 8, 105-122.

Siddiqi, M. A., (1995), An Indirect Test for Dividend Relevance, Journal of Financial

Research 18, 89-101.

Siems, T. F. (1996). Bank Mergers and Shareholder Wealth: Evidence from 1995s

Mega –merger Deals. Financial Industry Studies, Federal Reserve Bank of

Dallas, Texas.

Soter, D., Brigham, E. and Evanson, P. (1996), The Dividend Cut Heard' Round The

World: The Case of FPL, Journal of Applied Corporate Finance 9, 4-15.

Stevenson, C. & Filippi, I. (2004). How Does the European Commission‟s New Chief

Economist Think? European Competition Law Review, 25 (2), 122-125.

Subrahmanyam, V., Rangan, N. and Rosenstein, S. (1997). The role of outside

directors in bank acquisitions. Financial Management 26 (3),23-36.

Sudarsanma, S. & Mahate, A.A. (2003) Glamour Acquirers, Method of Payment and

Post- Acquisition Performance: The UK Evidence. Journal of Business Finance

& Accounting 30 (1&2), 299 -341.

Tabachnick, B.G. & Fidell, L.S. (2001). Using Multivariate Statistics.

London; Pearson Education.

The Economist 1997

The European Central Bank Report 2004

Tobin, J., (1969), A General Equilibrium Approach to Monetary Theory, Journal of

Money, Credit and Banking 1, 15-29.

Vafeas, N. (1999). Determinants of the Adoption of Director Incentive Plans. Journal

of Accounting, Auditing & Finance 14(4), Fall.

Valkanov, E. & Kleimeier, S. (2006). The Role of Regulatory Capital in International

Bank Mergers and Acquisitions. Research in International Business and Finance.

Vancil, R.F. (1987). Passing the Baton: Managing the Process of CEO Succession.

Harvard Business Press, Boston, MA.

207

Vander –Vennet, R. ( 1996) The Effect of Mergers and Acquisitions on the Efficiency

& Profitability of EC Credit Institutions. Journal of Banking & Finance, 20,

1531 -1558.

Vander-Vennet, R. (1998). Causes and consequences of EU bank takeovers. In :

Eijffinger, S., Koedijk, K., Pagano, M., Portes, R. (eds)., The Changing

European Landscape: Centre for Economic Policy Research, Brussels, 45-61.

Vermeeulen, F. & Barkema, H. (1996). Learning through Acquisitions. Academy of

Management Journal 44(3), 457 -476.

Walker, C. & Raes, J. (2005). Integration and Consolidation in EU Banking: An

Unfinished Business. Economic Papers Series, European Commission

BU1 –1/180.

Walker, R. L. (2005). Empirical analysis of up- or – out rules for promotion policies.

Journal of Economics and Finance 29 ( 2), 172 – 186.

Walsh, C. (2008). Key Management Ratios 4th

ed. London; Pearson Education.

Watson, D. & Head, A. (2005) Corporate Finance: Principles and Practice. New York;

Pearson Education.

Wei, J. G., Zhang,W. & Xiao, Z.J. (2003) Dividend payment and Ownership structure

in China. Advances in Financial Economics 9, 187–219.

Weisbach, M. S., (1988).Outside directors and CEO turnover. Journal of Financial

Economics 20 (1-2), 431 -460.

Wilkinson, B. R., Cahan, S.F. & Jones, G. (2001). Strategies and Dividend Imputation:

The Effect of Foreign and Domestic Ownership on Average Effective Tax Rates

.Journal of International Accounting, Auditing & Taxation 10(2), 157-175.

Wu, C. (1996). Taxes and Dividend Policy. International Review of Economics and

Finance, 5(3), 291-305.

Young, I. (2008). Credit Crunch Slows M&A in Europe. Chemical Week June 2, 24.

Yoon, P. S., Laura, T. & Starks, L. T. (1995). Signalling Investment Opportunities,

And Dividend Announcements. Review of Financial Studies 8, 995 -1018.

Zalan, T. & Lewis, G. (2005). The CEO – Advisors Nexus: Towards an Explanation

of Merger Preference in Mergers and Acquisitions. Problems & Perspective in

Management 4, 49-91.

Zhilan, C., Yan, L. C., Aris, S. & Anita, W. S. (2005) Ownership concentration, firm

performance, and dividend policy in Hong Kong. Pacific Basin Finance

Journal 13(4), 431 -449.

208

Zhou, K. & Wong, M. (2008). The Determinant of Net Interest Margins of

Commercial Banks in Mainland China. Emerging Markets Finance & Trade

44(5), 41-53.

Zollo, M. & Meier, D. (2008). What is M&A Performance? Academy of Management

Perspective 55 -77.

Appendix 1

List of Domestic Acquirers

Announcement Date Acquirers Acquirer country

1 16/12/2002 Sacam Developpement SAS [and others] France

2 15/11/2005 Commerzbank AG Germany

3 25/10/2005 Deutsche Postbank AG Germany

4 23/01/2006 Deutsche Postbank AG Germany

5 30/06/1999 Deutsche Postbank AG Germany

6 18/12/2001 Piraeus Bank SA Greece

7 23/06/2003 Piraeus Bank SA Greece

8 26/08/2006 Banca Intesa SpA Italy

9 20/05/2007 Unicredito Italiano SpA Italy

10 09/10/1998 Banca Monte dei Paschi di Siena SpA Italy

11 18/07/1997 Banca Popolare di Verona SpA Italy

12 25/03/2003 Sanpaolo IMI SpA Italy

13 15/07/2002 Unicredito Italiano SpA Italy

14 04/12/2006 Banche Popolari Unite Scpa Italy

15 03/01/2000 Skb Banka Dd Slovenia

16 19/02/1998 Banco Santander SA Spain

17 31/12/2000 Banco Santander Central Hispano SA Spain

18 27/06/2001 Banco Santander Central Hispano SA Spain

19 16/09/1997 Banco Santander SA Spain

20 26/01/2000 Royal Bank of Scotland Group PLC UK

Appendix 2

List of Cross Border Acquirers

S/N Announcement Date Acquirer Acquirer country

1 20/08/2005 Raiffeisen International Bank-Holding AG Austria

2 29/10/1999 Erste Bank der Oesterreichischen Sparkassen AG Austria

3 28/09/2007 KBC Groep NV Belgium

4 19/04/2002 KBC Bancassurance Holding SA Belgium

5 20/09/2006 Cyprus Popular Bank Public Company Ltd Cyprus

6 13/06/2006 Credit Agricole SA France

7 24/03/2006 Societe Generale France

8 14/05/2001 Societe Generale France

9 07/01/2001 BNP Paribas SA France

10 03/05/2004 Societe Generale France

11 14/01/2001 Societe Generale France

12 26/09/2002 Societe Generale France

13 03/03/1998 Societe Generale France

14 03/11/2003 Societe Generale France

15 22/07/2000 Bayerische Hypo-und Vereinsbank AG Germany

16 15/02/2002 Unicredito Italiano SpA [and others] Germany

17 09/11/2003 Commerzbank AG Germany

18 01/08/2007 National Bank of Greece SA Greece

19 06/12/2005 Unicredito Italiano SpA Italy

20 01/09/2007 Unicredito Italiano SpA Italy

21 23/03/2006 Unicredito Italiano SpA Italy

22 08/04/2004 Unicredito Italiano SpA Italy

23 06/12/2005 Unicredito Italiano SpA Italy

24 20/01/2003 Unicredito Italiano SpA Italy

25 15/06/2001 IntesaBci SpA Italy

26 04/01/2002 Unicredito Italiano SpA Italy

27 13/04/2000 Unicredito Italiano SpA Italy

28 21/03/2001 Unicredito Italiano SpA Italy

29 04/05/2006 Banco Comercial Portugues SA Portugal

30 23/07/2004 Banco Santander Central Hispano SA Spain

31 06/12/2006 Banco Bilbao Vizcaya SA Spain

32 29/12/2002 Banco Santander Central Hispano SA Spain

33 02/10/2000 Banco Santander Central Hispano SA Spain

34 06/12/2006 Banco Bilbao Vizcaya SA Spain

35 21/01/1999 Banco Santander Central Hispano SA - BSCH Spain

36 09/08/2005 Skandinaviska Enskilda Banken AB Sweden

37 23/09/2004 Barclays PLC UK

38 01/10/2005 Standard Chartered PLC UK

39 28/10/2003 HSBC Holdings Plc UK

40 17/07/2006 Standard Chartered PLC UK

41 20/09/2001 HSBC Holdings Plc UK

42 14/05/2001 Societe Generale France